Compare commits

...

139 Commits

Author SHA1 Message Date
Aiden Cline c562522e4f [opencode-go] Normalize DeepSeek V4 efforts 2026-06-11 23:57:47 -05:00
Aiden Cline 2d0173d177 [opencode-go] Add reasoning options 2026-06-11 23:56:33 -05:00
Aiden Cline 1a772fd297 Merge pull request #2223 from anomalyco/feat/google-vertex-reasoning-options-wave3
[google-vertex] Add MaaS reasoning options
2026-06-11 23:52:15 -05:00
Aiden Cline c63920050a [google-vertex] Add MaaS reasoning options 2026-06-11 23:50:13 -05:00
Aiden Cline df45483742 Merge pull request #2211 from anomalyco/feat/wafer.ai-reasoning-options-wave3
[wafer.ai] Add reasoning options
2026-06-11 23:49:22 -05:00
Aiden Cline 4ecdcb8f2b Merge pull request #2196 from anomalyco/feat/hpc-ai-reasoning-options-wave3
[hpc-ai] Add reasoning options
2026-06-11 23:48:29 -05:00
Aiden Cline 4644def1ae [hpc-ai] Remove unsupported reasoning efforts 2026-06-11 23:45:03 -05:00
Aiden Cline f9091af5ee Merge pull request #2198 from anomalyco/feat/mixlayer-reasoning-options-wave3
[mixlayer] Add reasoning options
2026-06-11 23:41:20 -05:00
Aiden Cline 7022a48bfc Merge pull request #2205 from anomalyco/feat/vultr-reasoning-options-wave3
[vultr] Add reasoning options
2026-06-11 23:40:45 -05:00
Aiden Cline dda69225d8 Merge pull request #2190 from anomalyco/feat/perplexity-reasoning-options-wave3
[perplexity] Add reasoning options
2026-06-11 23:38:49 -05:00
Aiden Cline 2e020b1dd2 Merge pull request #2194 from anomalyco/feat/minimax-coding-plan-reasoning-options-wave3
[minimax-coding-plan] Complete reasoning options
2026-06-11 23:38:32 -05:00
Aiden Cline 67252bda30 Merge pull request #2197 from anomalyco/feat/submodel-reasoning-options-wave3
[submodel] Add reasoning options
2026-06-11 23:38:20 -05:00
Aiden Cline ac1566f622 Merge pull request #2193 from anomalyco/feat/minimax-reasoning-options-wave3
[minimax] Complete reasoning options
2026-06-11 23:38:02 -05:00
Aiden Cline 48837609aa Merge pull request #2199 from anomalyco/feat/v0-reasoning-options-wave3
[v0] Add reasoning options
2026-06-11 23:37:49 -05:00
Aiden Cline c2a0ff023a Merge pull request #2138 from martinmose/add-zeldoc-provider
feat(provider): add zeldoc provider
2026-06-11 23:28:07 -05:00
Aiden Cline 282821e7b6 Merge pull request #2208 from anomalyco/feat/qihang-ai-reasoning-options-wave3
[qihang-ai] Add reasoning options
2026-06-11 23:25:07 -05:00
Aiden Cline 1b8e53bcbd Merge pull request #2192 from anomalyco/feat/minimax-cn-reasoning-options-wave3
[minimax-cn] Complete reasoning options
2026-06-11 23:14:09 -05:00
Aiden Cline f671147f71 Merge pull request #2207 from anomalyco/feat/scaleway-reasoning-options-wave3
[scaleway] Add reasoning options
2026-06-11 23:13:54 -05:00
Aiden Cline e23e759601 Merge pull request #2204 from anomalyco/feat/clarifai-reasoning-options-wave3
[clarifai] Add reasoning options
2026-06-11 23:13:40 -05:00
Aiden Cline 4b4546f3e3 Merge pull request #2213 from anomalyco/feat/friendli-reasoning-options-wave3
[friendli] Add reasoning options
2026-06-11 23:13:24 -05:00
Aiden Cline 040f0ca995 [wafer.ai] Add DeepSeek V4 effort controls 2026-06-11 23:12:29 -05:00
Aiden Cline 9acac34889 [friendli] Remove unsupported reasoning efforts 2026-06-11 23:11:37 -05:00
Aiden Cline 78ad91f773 Merge pull request #2212 from anomalyco/feat/iflowcn-reasoning-options-wave3
[iflowcn] Add reasoning options
2026-06-11 23:10:34 -05:00
Aiden Cline 4b0aeef538 Merge pull request #2214 from anomalyco/feat/io-net-reasoning-options-wave3
[io-net] Add reasoning options
2026-06-11 23:09:58 -05:00
Aiden Cline 856787cb84 [friendli] Add reasoning options 2026-06-11 23:08:36 -05:00
Aiden Cline 084f0bb4ba [iflowcn] Add reasoning options 2026-06-11 23:08:36 -05:00
Aiden Cline ee1301fb63 [io-net] Add reasoning options 2026-06-11 23:08:36 -05:00
Aiden Cline 502957b778 [wafer.ai] Add reasoning options 2026-06-11 23:08:29 -05:00
Aiden Cline 9f11a93d06 [vultr] Add reasoning options 2026-06-11 23:07:50 -05:00
Aiden Cline be8d8a2ec2 [qihang-ai] Add reasoning options 2026-06-11 23:07:50 -05:00
Aiden Cline d4193dbad6 [scaleway] Add reasoning options 2026-06-11 23:07:50 -05:00
Aiden Cline fb6b0985ce [clarifai] Add reasoning options 2026-06-11 23:07:50 -05:00
Aiden Cline 2bb4fe287e [hpc-ai] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline b58c396106 [mixlayer] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline 575f078887 [minimax-coding-plan] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline acf194ec01 [submodel] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline c80ff1ad2c [minimax] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline a41f6f5913 [v0] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline 8dffe03a9c [minimax-cn] Add reasoning options 2026-06-11 23:07:20 -05:00
Aiden Cline 9ea761af1a [perplexity] Add reasoning options 2026-06-11 23:07:07 -05:00
Aiden Cline 526b4cd543 Merge pull request #2179 from anomalyco/feat/deepseek-reasoning-options-wave2
[deepseek] Complete reasoning options
2026-06-11 22:55:29 -05:00
Aiden Cline b7d165296c Merge pull request #2172 from anomalyco/feat/privatemode-ai-reasoning-options
[privatemode-ai] Add reasoning options
2026-06-11 22:55:21 -05:00
Aiden Cline 655f757925 Merge pull request #2177 from anomalyco/feat/llmtr-reasoning-options
[llmtr] Complete reasoning options
2026-06-11 22:55:07 -05:00
Aiden Cline cce129dfd9 Merge pull request #2173 from anomalyco/feat/zai-coding-plan-reasoning-options
[zai-coding-plan] Complete reasoning options
2026-06-11 22:54:57 -05:00
Aiden Cline 548f3b5869 Merge pull request #2175 from anomalyco/feat/zhipuai-coding-plan-reasoning-options
[zhipuai-coding-plan] Complete reasoning options
2026-06-11 22:54:51 -05:00
Aiden Cline f349a9dde6 Merge pull request #2176 from anomalyco/feat/kuae-cloud-reasoning-options
[kuae-cloud-coding-plan] Add reasoning options
2026-06-11 22:54:34 -05:00
Aiden Cline 05c1041160 [deepseek] Mark reasoner controls fixed 2026-06-11 22:54:19 -05:00
Aiden Cline 215c3dfbc8 Merge pull request #2174 from anomalyco/feat/kimi-for-coding-reasoning-options
[kimi-for-coding] Complete reasoning options
2026-06-11 22:54:16 -05:00
Aiden Cline ddc6ad75a2 Merge pull request #2178 from anomalyco/feat/firepass-reasoning-options
[firepass] Add reasoning options
2026-06-11 22:54:05 -05:00
Aiden Cline 757bee7de8 Merge pull request #2167 from anomalyco/feat/claudinio-reasoning-options
[claudinio] Add reasoning options
2026-06-11 22:53:39 -05:00
Aiden Cline 5ee32a47ae Merge pull request #2170 from anomalyco/feat/bailing-reasoning-options
[bailing] Add reasoning options
2026-06-11 22:53:30 -05:00
Aiden Cline 2c22f4b488 [privatemode-ai] Add reasoning options 2026-06-11 22:51:42 -05:00
Aiden Cline 097296f6c6 [llmtr] Add reasoning options 2026-06-11 22:51:42 -05:00
Aiden Cline 2559ed64c5 [zai-coding-plan] Add reasoning options 2026-06-11 22:51:42 -05:00
Aiden Cline c56ac404d3 [zhipuai-coding-plan] Add reasoning options 2026-06-11 22:51:42 -05:00
Aiden Cline 4681e30afb [kuae-cloud-coding-plan] Add reasoning options 2026-06-11 22:51:42 -05:00
Aiden Cline 94cd023a88 [kimi-for-coding] Add reasoning options 2026-06-11 22:51:42 -05:00
Aiden Cline b5c33b6347 [deepseek] Add reasoning options 2026-06-11 22:51:41 -05:00
Aiden Cline 2014d883e4 [firepass] Add reasoning options 2026-06-11 22:51:41 -05:00
Aiden Cline 2f749b7cf9 [claudinio] Add reasoning options 2026-06-11 22:51:31 -05:00
Aiden Cline d7ab976e3c [bailing] Add reasoning options 2026-06-11 22:51:31 -05:00
Aiden Cline 32066b7856 Merge pull request #2131 from anomalyco/feat/ovhcloud-reasoning-audit
feat(ovhcloud): add reasoning options
2026-06-11 22:42:30 -05:00
Aiden Cline 3ded2fae72 Merge pull request #2130 from anomalyco/audit/nebius-models
[nebius] Audit reasoning controls
2026-06-11 22:38:33 -05:00
Aiden Cline 9771180f83 [nebius] Correct reasoning controls 2026-06-11 22:31:54 -05:00
Aiden Cline fa65113f37 Merge pull request #2162 from mikeyp/chore/update-digitalocean-models
Add anthropic-claude-fable-5 and nemotron-3-ultra-550b to DigitalOcean
2026-06-11 22:19:22 -05:00
Aiden Cline 1ab5784118 Merge pull request #2161 from andrelandgraf/feat/add-neon-provider
Add Neon provider
2026-06-11 22:17:07 -05:00
Mike Prasuhn 411bc157e6 Add anthropic-claude-fable-5 and nemotron-3-ultra-550b to DigitalOcean 2026-06-11 23:09:08 -04:00
Andre Landgraf 58ab76b8f3 Add Neon provider
Neon serves the same Databricks-backed model catalog through its
branch-scoped AI Gateway via an OpenAI-compatible endpoint, so this mirrors
the `databricks` provider's models.

- `api` uses the branch-scoped `NEON_AI_GATEWAY_BASE_URL` + the unified MLflow
  OpenAI-compatible route; `NEON_AI_GATEWAY_TOKEN` is the bearer key. Both are
  emitted by `neonctl env pull`.
- Model ids drop the `databricks-` prefix (the gateway accepts the bare ids),
  so models resolve as `neon/claude-haiku-4-5`, `neon/gpt-5-nano`, etc.
2026-06-11 19:52:37 -07:00
Martin Mose Facondini 8f2607fcdb fix(zeldoc): make logo black 2026-06-12 00:12:49 +02:00
Martin Mose Facondini 23e07a27d2 fix(zeldoc): correct z-code model fields 2026-06-12 00:12:44 +02:00
Aiden Cline 37e8e0cf95 Merge pull request #2086 from anomalyco/feat/openrouter-reasoning-options
feat(openrouter): add reasoning options
2026-06-11 16:21:15 -05:00
Aiden Cline 0e0fe311ab Merge dev into feat/openrouter-reasoning-options 2026-06-11 16:14:33 -05:00
Aiden Cline 3d763d081e fix(openrouter): document Claude effort mapping 2026-06-11 16:14:03 -05:00
Aiden Cline ebbd3416e2 Merge pull request #2154 from anomalyco/automation/sync-models-openrouter
chore(sync): update OpenRouter model catalog
2026-06-11 16:07:40 -05:00
Aiden Cline a3509f9a3d Merge pull request #2155 from anomalyco/automation/sync-models-vercel
chore(sync): update Vercel AI Gateway model catalog
2026-06-11 16:07:24 -05:00
github-actions[bot] 71d00e7dfc chore(sync): update Vercel AI Gateway model catalog 2026-06-11 21:06:24 +00:00
github-actions[bot] 26f7b6f6b6 chore(sync): update OpenRouter model catalog 2026-06-11 21:06:21 +00:00
Aiden Cline 270009bd93 fix(openrouter): correct reasoning controls 2026-06-11 15:03:25 -05:00
Aiden Cline 91cc389af3 fix(openrouter): correct Claude reasoning controls 2026-06-11 14:57:34 -05:00
Aiden Cline 1c7b7e8247 Merge dev into feat/openrouter-reasoning-options 2026-06-11 14:27:05 -05:00
Aiden Cline a382026ce9 Merge pull request #2153 from anomalyco/automation/sync-models-vercel
chore(sync): update Vercel AI Gateway model catalog
2026-06-11 14:21:25 -05:00
github-actions[bot] 1fc7261c14 chore(sync): update Vercel AI Gateway model catalog 2026-06-11 19:15:03 +00:00
Aiden Cline 765dae9f61 Merge pull request #2134 from anomalyco/audit/deepinfra-reasoning
fix(deepinfra): reconcile reasoning controls
2026-06-11 13:02:22 -05:00
Aiden Cline 02cd80a2ab Merge pull request #2032 from anthraxx/alibaba-qwen3.7-plus
add Qwen3.7 Plus model configuration to Alibana coding plan
2026-06-11 12:57:58 -05:00
Aiden Cline fcbd02fa84 Merge pull request #2148 from anomalyco/automation/sync-models-venice
chore(sync): update Venice model catalog
2026-06-11 12:56:57 -05:00
Aiden Cline e07a0bab97 Merge pull request #2132 from anomalyco/chore/fireworks-reasoning
fix(fireworks-ai): reconcile reasoning controls
2026-06-11 12:55:12 -05:00
Aiden Cline ad166f7448 fix(fireworks-ai): verify reasoning toggles 2026-06-11 12:33:12 -05:00
github-actions[bot] 660350c5fa chore(sync): update Venice model catalog 2026-06-11 17:29:06 +00:00
Aiden Cline 9379be8911 Merge pull request #2151 from anomalyco/fix/togetherai-required-reasoning-options
[togetherai] Require reasoning options metadata
2026-06-11 12:21:18 -05:00
Aiden Cline 1ccb247f7f [togetherai] Require reasoning options metadata 2026-06-11 12:05:22 -05:00
Aiden Cline 7d5469898d [nebius] Complete reasoning option coverage 2026-06-11 12:04:48 -05:00
Aiden Cline ec3c4ed8ea fix(fireworks-ai): mark unresolved reasoning controls 2026-06-11 12:04:47 -05:00
Aiden Cline 8521822a96 Merge pull request #2135 from anomalyco/audit/togetherai-reasoning-20260610
Audit Together AI models and reasoning controls
2026-06-11 12:01:15 -05:00
Aiden Cline 9df50e0ccc chore(ovhcloud): remove test changes 2026-06-11 12:00:37 -05:00
Aiden Cline 9b3d25aae8 [nebius] Remove provider matrix test 2026-06-11 12:00:36 -05:00
Aiden Cline ea4d10b219 chore(fireworks-ai): remove catalog test 2026-06-11 12:00:36 -05:00
Aiden Cline d3d163dd95 Remove Together provider matrix test 2026-06-11 12:00:35 -05:00
Aiden Cline 78f8fb92fc Merge pull request #2129 from anomalyco/audit/cerebras-models-20260610
[cerebras] Refresh public model catalog
2026-06-11 12:00:23 -05:00
Aiden Cline 62b4296b3d Delete packages/core/test/cerebras.test.ts 2026-06-11 12:00:10 -05:00
Aiden Cline 12057804f5 Merge pull request #2143 from Nindaleth/feature/ghcp-fable
feat(github-copilot): add Claude Fable 5 model
2026-06-11 11:48:42 -05:00
Aiden Cline 127290ebec Merge pull request #2142 from anomalyco/automation/sync-models-openrouter
chore(sync): update OpenRouter model catalog
2026-06-11 11:45:55 -05:00
Aiden Cline 5abbfce7d6 Merge pull request #2120 from davidfierro/feat/snowflake-cortex-models
feat(snowflake-cortex): add missing but officially supported models
2026-06-11 11:45:36 -05:00
Aiden Cline 6ff6ab4028 Merge pull request #2145 from Omee11/feat/token-plan-cn
feat(alibaba-token-plan-cn): add Alibaba Token Plan (China) provider
2026-06-11 11:36:16 -05:00
Aiden Cline 0f3e0d7337 Merge pull request #2146 from Omee11/feat/token-plan-qwen3.7-plus
feat(alibaba-token-plan): add qwen3.7-plus
2026-06-11 10:51:52 -05:00
github-actions[bot] 8381089b11 chore(sync): update OpenRouter model catalog 2026-06-11 15:28:27 +00:00
Oliver Mee 98bed4baf6 feat(alibaba-token-plan-cn): add Alibaba Token Plan (China) provider 2026-06-11 18:14:20 +08:00
Oliver Mee e55ab2daf8 feat(alibaba-token-plan): add qwen3.7-plus 2026-06-11 18:14:20 +08:00
Radek Liska c60784fab5 feat(github-copilot): add Claude Fable 5 model 2026-06-11 09:15:02 +02:00
Aiden Cline 21581d8f4c test(sync): cover factored reasoning overrides 2026-06-10 23:20:55 -05:00
Aiden Cline 20bfe37a53 fix(sync): resolve changed canonical base 2026-06-10 23:19:32 -05:00
Aiden Cline fc4a781c2f fix(ovhcloud): complete reasoning controls 2026-06-10 23:16:08 -05:00
Aiden Cline 28674e1af4 [cerebras] Assert complete resolved models 2026-06-10 23:16:01 -05:00
Aiden Cline 1e76995f8e Test resolved Together provider matrix 2026-06-10 23:15:34 -05:00
Aiden Cline 0e371b761d fix(sync): resolve reasoning before preservation 2026-06-10 23:14:13 -05:00
Aiden Cline 7a5bf4f56e [cerebras] Test resolved model matrix 2026-06-10 23:13:39 -05:00
Aiden Cline 4c5b17b1db [nebius] Test generated provider matrix 2026-06-10 23:13:35 -05:00
Aiden Cline c4650219c3 fix(sync): drop stale reasoning options 2026-06-10 23:13:06 -05:00
Aiden Cline 9fb0474d1c feat(ovhcloud): add reasoning options 2026-06-10 23:13:06 -05:00
Aiden Cline 8807bb0069 Correct Together reasoning and pricing metadata 2026-06-10 23:12:12 -05:00
Aiden Cline 18c35709c5 Merge pull request #2139 from anomalyco/fix/venice-sync-models
Venice: fix synced model metadata
2026-06-10 20:14:38 -05:00
Aiden Cline 6799ff1078 [nebius] Correct verified reasoning controls 2026-06-10 20:02:58 -05:00
Aiden Cline 236ff0e39b fix(fireworks-ai): add Qwen reasoning budget 2026-06-10 20:02:44 -05:00
Aiden Cline cb4fd81c59 Correct Together Qwen reasoning metadata 2026-06-10 19:51:01 -05:00
Aiden Cline 03c161f038 [cerebras] Correct GLM reasoning option 2026-06-10 19:50:19 -05:00
Aiden Cline 38a2f09999 [nebius] Reconcile model lifecycle evidence 2026-06-10 19:49:42 -05:00
Aiden Cline 1f77766834 fix(fireworks-ai): remove unverified toggles 2026-06-10 19:48:27 -05:00
Aiden Cline da1032a1cb [venice] Fix synced model metadata 2026-06-10 19:46:52 -05:00
Aiden Cline 55848d41c6 Merge pull request #2123 from BlockListed/cortecs-add-claude-opus-4-8
add claude opus 4.8 to cortecs
2026-06-10 19:36:26 -05:00
BlockListed e8304a0b0f add claude opus 4.8 to cortecs 2026-06-10 23:48:02 +02:00
Martin Mose Facondini 23b4754d23 rename agentic-coding model to z-code 2026-06-10 23:06:51 +02:00
Martin Mose Facondini 20ffc3909f add zeldoc provider with agentic-coding model 2026-06-10 23:06:21 +02:00
Aiden Cline 09c7f864f2 Audit Together AI model catalog and reasoning 2026-06-10 16:05:11 -05:00
Aiden Cline 5a3e0cacea test(fireworks-ai): lock reasoning controls 2026-06-10 16:04:34 -05:00
Aiden Cline f8ccb57731 [nebius] Audit reasoning controls 2026-06-10 16:04:09 -05:00
Aiden Cline c0b03ed655 [cerebras] Refresh public model catalog 2026-06-10 16:03:52 -05:00
David Fierro Iglesias 5dcd077370 feat(snowflake-cortex): add officially supported models 2026-06-10 16:29:41 +02:00
Aiden Cline 5e7769cb71 fix(openrouter): expose Claude Opus effort 2026-06-08 20:22:10 -05:00
Aiden Cline add3daacea feat(openrouter): add reasoning options 2026-06-08 20:17:16 -05:00
Levente Polyak 15f015fd4a add Qwen3.7 Plus model configuration to Alibana coding plan
Coding-plan models are at a fixed monthly fee.

Link: https://modelstudio.console.alibabacloud.com/eu-central-1?tab=doc#/doc/?type=model&url=3005961
2026-06-07 21:54:13 +02:00
553 changed files with 2136 additions and 1035 deletions
-17
View File
@@ -1,17 +0,0 @@
name = "Aion 2.0"
family = "o"
release_date = "2026-03-24"
last_updated = "2026-06-10"
attachment = false
reasoning = true
tool_call = false
temperature = true
open_weights = false
[limit]
context = 128000
output = 32768
[modalities]
input = ["text"]
output = ["text"]
@@ -1,21 +0,0 @@
name = "Trinity Large Thinking"
family = "trinity"
release_date = "2026-04-02"
last_updated = "2026-06-10"
attachment = false
reasoning = true
tool_call = true
structured_output = true
temperature = true
open_weights = true
[limit]
context = 256000
output = 65536
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/arcee-ai/Trinity-Large-Thinking-FP8-Block"
-20
View File
@@ -1,20 +0,0 @@
name = "Gemma 3 27B"
family = "gemma"
release_date = "2026-03-18"
last_updated = "2026-06-10"
attachment = false
reasoning = false
tool_call = false
temperature = true
open_weights = true
[limit]
context = 40000
output = 4096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/google/gemma-3-27b-it"
-20
View File
@@ -1,20 +0,0 @@
name = "Gemma 4 31B Instruct"
family = "gemma"
release_date = "2026-05-20"
last_updated = "2026-06-10"
attachment = false
reasoning = true
tool_call = false
temperature = true
open_weights = true
[limit]
context = 32000
output = 4096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/google/gemma-4-31B-it"
-20
View File
@@ -1,20 +0,0 @@
name = "Qwen 2.5 7B"
family = "qwen"
release_date = "2026-03-18"
last_updated = "2026-06-10"
attachment = false
reasoning = false
tool_call = false
temperature = true
open_weights = true
[limit]
context = 32000
output = 4096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen2.5-7B-Instruct"
-20
View File
@@ -1,20 +0,0 @@
name = "Qwen3 30B A3B"
family = "qwen"
release_date = "2026-03-18"
last_updated = "2026-06-10"
attachment = false
reasoning = false
tool_call = true
temperature = true
open_weights = true
[limit]
context = 256000
output = 32768
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-30B-A3B-Instruct-2507"
@@ -1,17 +0,0 @@
name = "Qwen3.6 35B A3B Uncensored"
family = "qwen3.6"
release_date = "2026-05-24"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false
[limit]
context = 128_000
output = 4_096
[modalities]
input = ["text"]
output = ["text"]
-20
View File
@@ -1,20 +0,0 @@
name = "Qwen 3.6 35B A3B FP8"
family = "qwen"
release_date = "2026-05-20"
last_updated = "2026-06-10"
attachment = false
reasoning = true
temperature = true
tool_call = true
open_weights = true
[limit]
context = 32_000
output = 4_096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3.6-35B-A3B-FP8"
@@ -1,20 +0,0 @@
name = "Qwen3 VL 30B A3B"
family = "qwen"
release_date = "2026-03-18"
last_updated = "2026-06-10"
attachment = true
reasoning = false
tool_call = true
temperature = true
open_weights = true
[limit]
context = 128000
output = 4096
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-VL-30B-A3B-Instruct"
@@ -1,20 +0,0 @@
name = "Venice Uncensored 1.1"
family = "venice"
release_date = "2026-03-18"
last_updated = "2026-06-10"
attachment = false
reasoning = false
tool_call = false
temperature = true
open_weights = true
[limit]
context = 32000
output = 4096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition"
-21
View File
@@ -1,21 +0,0 @@
name = "Gemma 4 Uncensored"
family = "gemma"
release_date = "2026-04-13"
last_updated = "2026-06-10"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 256_000
output = 8_192
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Jiunsong/supergemma4-26b-uncensored-gguf-v2"
-22
View File
@@ -1,22 +0,0 @@
name = "Google Gemma 3 27B Instruct"
family = "gemma"
release_date = "2025-11-04"
last_updated = "2026-06-10"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-07"
open_weights = true
[limit]
context = 198_000
output = 16_384
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/google/gemma-3-27b-it"
-18
View File
@@ -1,18 +0,0 @@
name = "Grok 4.20 Multi-Agent"
family = "grok"
release_date = "2026-03-12"
last_updated = "2026-06-10"
attachment = true
reasoning = true
temperature = true
tool_call = false
structured_output = true
open_weights = false
[limit]
context = 2_000_000
output = 128_000
[modalities]
input = ["text", "image"]
output = ["text"]
@@ -1,21 +0,0 @@
name = "Hermes 3 Llama 3.1 405b"
family = "hermes"
release_date = "2025-09-25"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = false
knowledge = "2024-04"
open_weights = true
[limit]
context = 128_000
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/NousResearch/Hermes-3-Llama-3.1-405B"
-21
View File
@@ -1,21 +0,0 @@
name = "Llama 3.2 3B"
family = "llama"
release_date = "2024-10-03"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2023-12"
open_weights = true
[limit]
context = 128_000
output = 4_096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/meta-llama/Llama-3.2-3B"
-21
View File
@@ -1,21 +0,0 @@
name = "Llama 3.3 70B"
family = "llama"
release_date = "2025-04-06"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = true
knowledge = "2023-12"
open_weights = true
[limit]
context = 128_000
output = 4_096
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/meta-llama/Llama-3.3-70B-Instruct"
-18
View File
@@ -1,18 +0,0 @@
name = "Mercury 2"
family = "mercury"
release_date = "2026-02-20"
last_updated = "2026-06-10"
attachment = false
reasoning = true
tool_call = true
structured_output = true
temperature = true
open_weights = false
[limit]
context = 128000
output = 50000
[modalities]
input = ["text"]
output = ["text"]
@@ -1,21 +0,0 @@
name = "Mistral Small 3.2 24B Instruct"
family = "mistral-small"
release_date = "2026-01-15"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 256_000
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506"
@@ -1,21 +0,0 @@
name = "GLM 4.7 Flash Heretic"
family = "glm"
release_date = "2026-02-04"
last_updated = "2026-06-10"
attachment = false
reasoning = true
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 200_000
output = 24_000
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Olafangensan/GLM-4.7-Flash-heretic"
-21
View File
@@ -1,21 +0,0 @@
name = "OpenAI GPT OSS 120B"
family = "gpt-oss"
release_date = "2025-11-06"
last_updated = "2026-06-10"
attachment = false
reasoning = true
tool_call = true
temperature = true
knowledge = "2025-07"
open_weights = true
[limit]
context = 128000
output = 16384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/openai/gpt-oss-120b"
@@ -1,22 +0,0 @@
name = "Qwen 3 235B A22B Instruct 2507"
family = "qwen"
release_date = "2025-04-29"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-07"
open_weights = true
[limit]
context = 128_000
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-235B-A22B-Instruct-2507-FP8"
@@ -1,22 +0,0 @@
name = "Qwen 3 235B A22B Thinking 2507"
family = "qwen"
release_date = "2025-04-29"
last_updated = "2026-06-10"
attachment = false
reasoning = true
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-07"
open_weights = true
[limit]
context = 128_000
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-235B-A22B-Thinking-2507-FP8"
@@ -1,21 +0,0 @@
name = "Qwen 3 Coder 480B Turbo"
family = "qwen"
release_date = "2026-01-27"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 256_000
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo"
-22
View File
@@ -1,22 +0,0 @@
name = "Qwen 3 Next 80b"
family = "qwen"
release_date = "2025-04-29"
last_updated = "2026-06-10"
attachment = false
reasoning = false
temperature = true
tool_call = true
structured_output = true
knowledge = "2025-07"
open_weights = true
[limit]
context = 256_000
output = 16_384
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-Next-80B-A3B-Instruct"
-21
View File
@@ -1,21 +0,0 @@
name = "Qwen3 VL 235B"
family = "qwen3.5"
release_date = "2026-01-16"
last_updated = "2026-06-10"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 256_000
output = 16_384
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/Qwen/Qwen3-VL-235B-A22B-Instruct"
-21
View File
@@ -1,21 +0,0 @@
name = "Venice Uncensored 1.2"
family = "venice"
release_date = "2026-04-01"
last_updated = "2026-06-10"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 128_000
output = 8_192
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/cognitivecomputations/Dolphin-Mistral-24B-Venice-Edition"
@@ -1,21 +0,0 @@
name = "Venice Role Play Uncensored"
family = "venice"
release_date = "2026-02-20"
last_updated = "2026-06-10"
attachment = true
reasoning = false
temperature = true
tool_call = true
structured_output = true
open_weights = true
[limit]
context = 128_000
output = 4_096
[modalities]
input = ["text", "image"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/dphnAI/24B-3.2-RP-K2-final"
+28 -3
View File
@@ -51,6 +51,7 @@ export interface SyncProvider<SourceModel> {
skipCreates?: boolean;
deleteMissing?: boolean;
preserveSymlinks?: boolean;
preserveBaseModels?: boolean;
sameModel?(current: ExistingModel, desired: SyncedModel): boolean;
missingNotice?(paths: string[]): string[];
sourceID?(model: SourceModel): string;
@@ -118,7 +119,9 @@ export async function syncProvider<SourceModel>(
): Promise<SyncResult> {
console.log(`\nSyncing ${provider.name}...`);
const { models: existing, brokenSymlinks } = await readExisting(provider.modelsDir);
const existingState = await readExisting(provider.modelsDir);
const { models: existing, brokenSymlinks } = existingState;
let { modelMetadata } = existingState;
const sourceModels = provider.parseModels(await provider.fetchModels());
const desired = new Map<string, { model: z.infer<typeof SyncedAuthoredModel>; content: string }>();
const desiredMetadata = new Map<string, { model: z.infer<typeof ModelMetadata>; content: string }>();
@@ -162,11 +165,28 @@ export async function syncProvider<SourceModel>(
});
}
const translatedModel = provider.preserveBaseModels === false
? translated.model
: preserveBaseModel(translated.model, existing.get(relativePath)?.authored);
const translatedBase = "base_model" in translatedModel ? translatedModel.base_model : undefined;
let resolvedReasoning: boolean | undefined;
if (translatedBase !== undefined) {
if (translated.metadata?.id === translatedBase) {
resolvedReasoning = translated.metadata.model.reasoning;
} else {
modelMetadata ??= await readModelMetadata(provider.modelsDir);
const canonicalReasoning = modelMetadata[translatedBase]?.reasoning;
resolvedReasoning = typeof canonicalReasoning === "boolean" ? canonicalReasoning : undefined;
}
} else {
resolvedReasoning = existing.get(relativePath)?.toml.reasoning;
}
const parsed = SyncedAuthoredModel.safeParse(stripUndefined({
id: translated.id,
...preserveReasoningOptions(
preserveBaseModel(translated.model, existing.get(relativePath)?.authored),
translatedModel,
existing.get(relativePath)?.authored,
resolvedReasoning,
),
}));
if (!parsed.success) {
@@ -316,7 +336,12 @@ export function preserveBaseModel(model: SyncedModel, existing: ExistingModel |
export function preserveReasoningOptions(
model: SyncedModel,
existing: ExistingModel | undefined,
resolvedReasoning: boolean | undefined = existing?.reasoning,
): SyncedModel {
if ((model.reasoning ?? resolvedReasoning) === false) {
const { reasoning_options: _reasoningOptions, ...withoutReasoningOptions } = model;
return withoutReasoningOptions as SyncedModel;
}
if (model.reasoning_options !== undefined || existing?.reasoning_options === undefined) return model;
return {
...model,
@@ -389,7 +414,7 @@ async function readExisting(modelsDir: string) {
existing.set(file, { authored, toml, symlink });
}
return { models: existing, brokenSymlinks };
return { models: existing, brokenSymlinks, modelMetadata };
}
async function isSymlink(filePath: string) {
@@ -121,6 +121,7 @@ export function buildOvhcloudModel(
last_updated: lastUpdated,
attachment,
reasoning,
reasoning_options: reasoning ? existing?.reasoning_options : undefined,
temperature: temperature || undefined,
tool_call: toolCall,
structured_output: structuredOutput || undefined,
+8 -35
View File
@@ -3,7 +3,7 @@ import path from "node:path";
import { z } from "zod";
import { ModelFamilyValues } from "../../family.js";
import type { ExistingModel, SyncProvider, SyncedFullModel, SyncedMetadata, SyncedModel } from "../index.js";
import type { ExistingModel, SyncProvider, SyncedFullModel, SyncedModel } from "../index.js";
import { factorBaseModel } from "./openrouter.js";
const API_ENDPOINT = "https://api.venice.ai/api/v1/models?type=text";
@@ -11,6 +11,7 @@ const MODELS_DIR = path.join(import.meta.dirname, "..", "..", "..", "..", "..",
const Capabilities = z.object({
supportsAudioInput: z.boolean().optional(),
supportsE2EE: z.boolean().optional(),
supportsFunctionCalling: z.boolean().optional(),
supportsReasoning: z.boolean().optional(),
supportsReasoningEffort: z.boolean().optional(),
@@ -82,7 +83,7 @@ export const venice = {
id: "venice",
name: "Venice",
modelsDir: "providers/venice/models",
metadataNamespace: "venice",
preserveBaseModels: false,
async fetchModels() {
const headers = process.env.VENICE_API_KEY
? { Authorization: `Bearer ${process.env.VENICE_API_KEY}` }
@@ -97,20 +98,14 @@ export const venice = {
return VeniceResponse.parse(raw).data;
},
translateModel(model, context) {
if (model.model_spec.capabilities.supportsE2EE === true) return undefined;
const id = model.id.replaceAll("/", "-");
const existing = context.existing(id);
const resolvedBase = existing?.base_model ?? resolveVeniceBaseModel(model.id, model.model_spec.name);
const baseModel = resolvedBase ?? `venice/${id}`;
const full = buildVeniceModel(model, existing, null);
const metadata = baseModel.startsWith("venice/")
? { id: baseModel, model: buildVeniceMetadata(full, model.model_spec.modelSource) }
: undefined;
const existingBase = existing?.base_model?.startsWith("venice/") === false ? existing.base_model : undefined;
const resolvedBase = existingBase ?? resolveVeniceBaseModel(model.id, model.model_spec.name);
return {
id,
model: resolvedBase === undefined
? factorNewMetadata(baseModel, full)
: buildVeniceModel(model, existing, baseModel),
metadata,
model: buildVeniceModel(model, existing, resolvedBase ?? null),
};
},
} satisfies SyncProvider<VeniceModel>;
@@ -166,7 +161,7 @@ export function buildVeniceModel(
reasoning_options: reasoningOptions,
tool_call: capabilities.supportsFunctionCalling === true,
structured_output: capabilities.supportsResponseSchema === true ? true : undefined,
temperature: true,
temperature: undefined,
cost,
limit,
modalities: { input: [...new Set(input)], output: ["text" as const] },
@@ -245,28 +240,6 @@ function inferFamily(id: string, name: string) {
});
}
function buildVeniceMetadata(model: SyncedModel, source: string | undefined): SyncedMetadata {
if ("base_model" in model) throw new Error("Cannot build Venice metadata from a factored model");
const { cost: _cost, reasoning_options: _reasoningOptions, interleaved: _interleaved, status: _status, ...metadata } = model;
return {
...metadata,
weights: model.open_weights && source?.startsWith("https://huggingface.co/")
? [{ url: source }]
: undefined,
};
}
function factorNewMetadata(baseModel: string, model: SyncedModel): SyncedModel {
if ("base_model" in model) return model;
return {
base_model: baseModel,
reasoning_options: model.reasoning_options,
cost: model.cost,
status: model.status,
interleaved: model.interleaved,
};
}
function stable(value: unknown): string {
if (Array.isArray(value)) return `[${value.map(stable).sort().join(",")}]`;
if (value !== null && typeof value === "object") {
+27 -3
View File
@@ -5,6 +5,7 @@ import path from "node:path";
import {
buildVeniceModel,
resolveVeniceBaseModel,
venice,
VeniceResponse,
type VeniceModel,
} from "../src/sync/providers/venice.js";
@@ -61,6 +62,25 @@ test("Venice emits empty reasoning options when efforts are unavailable", () =>
expect(synced).toMatchObject({ reasoning: true, reasoning_options: [] });
});
test("Venice does not infer temperature support", () => {
const synced = buildVeniceModel(catalogModel, undefined, null, "2026-06-10");
expect(synced.temperature).toBeUndefined();
});
test("Venice skips E2EE models", () => {
const translated = venice.translateModel({
...catalogModel,
id: "e2ee-test-model",
model_spec: {
...catalogModel.model_spec,
capabilities: { ...catalogModel.model_spec.capabilities, supportsE2EE: true },
},
}, { existing: () => undefined });
expect(translated).toBeUndefined();
});
test("Venice uses boundary-aware family matching", () => {
const synced = buildVeniceModel({
...catalogModel,
@@ -108,6 +128,7 @@ test("Venice maps API fields and keeps inherited models compact", () => {
expect(synced).not.toHaveProperty("release_date");
expect(synced).not.toHaveProperty("open_weights");
expect(synced).not.toHaveProperty("modalities");
expect(synced).not.toHaveProperty("temperature");
});
test("Venice preserves last_updated when authoritative data is unchanged", () => {
@@ -127,7 +148,7 @@ test("Venice rejects malformed responses", () => {
expect(() => VeniceResponse.parse({ data: [{ id: "broken" }] })).toThrow();
});
test("all Venice models use metadata inheritance and declare reasoning options", async () => {
test("Venice models use only canonical metadata and declare reasoning options", async () => {
const root = path.join(import.meta.dirname, "..", "..", "..");
const modelsDir = path.join(root, "providers", "venice", "models");
@@ -136,8 +157,11 @@ test("all Venice models use metadata inheritance and declare reasoning options",
base_model?: string;
reasoning_options?: unknown[];
};
expect(model.base_model, file).toBeDefined();
expect(model.reasoning_options, file).toBeDefined();
expect(await Bun.file(path.join(root, "models", `${model.base_model}.toml`)).exists(), file).toBe(true);
if (model.base_model !== undefined) {
expect(model.base_model.startsWith("venice/"), file).toBe(false);
expect(await Bun.file(path.join(root, "models", `${model.base_model}.toml`)).exists(), file).toBe(true);
}
expect(file.startsWith("e2ee-"), file).toBe(false);
}
});
@@ -0,0 +1,7 @@
base_model = "alibaba/qwen3.7-plus"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
@@ -0,0 +1,15 @@
base_model = "minimax/MiniMax-M2.5"
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
[limit]
context = 196_608
input = 196_601
output = 24_576
@@ -1,22 +1,23 @@
name = "DeepSeek V3.2"
family = "deepseek"
release_date = "2025-12-04"
last_updated = "2026-06-10"
release_date = "2025-12-03"
last_updated = "2025-12-05"
attachment = false
reasoning = true
tool_call = true
structured_output = true
temperature = true
knowledge = "2025-10"
tool_call = true
knowledge = "2025-01"
open_weights = true
[cost]
input = 0
output = 0
[limit]
context = 160000
output = 32768
context = 131_072
output = 65_536
[modalities]
input = ["text"]
output = ["text"]
[[weights]]
url = "https://huggingface.co/deepseek-ai/DeepSeek-V3.2"
output = ["text"]
@@ -0,0 +1,9 @@
base_model = "deepseek/deepseek-v4-flash"
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
@@ -0,0 +1,9 @@
base_model = "deepseek/deepseek-v4-pro"
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
@@ -0,0 +1,14 @@
base_model = "zhipuai/glm-5.1"
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
[limit]
context = 202_752
output = 128_000
@@ -0,0 +1,15 @@
base_model = "zhipuai/glm-5"
open_weights = false
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
[limit]
context = 202_752
output = 16_384
@@ -0,0 +1,17 @@
base_model = "moonshotai/kimi-k2.5"
base_model_omit = ["structured_output"]
family = "kimi"
attachment = true
temperature = true
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
[limit]
output = 32_768
@@ -0,0 +1,15 @@
base_model = "moonshotai/kimi-k2.6"
base_model_omit = ["structured_output"]
family = "kimi"
[interleaved]
field = "reasoning_content"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
[limit]
output = 16_384
@@ -0,0 +1,21 @@
name = "Qwen Image 2.0 Pro"
family = "qwen"
release_date = "2026-03-03"
last_updated = "2026-03-03"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false
[cost]
input = 0
output = 0
[limit]
context = 8_192
output = 0
[modalities]
input = ["text"]
output = ["image"]
@@ -0,0 +1,21 @@
name = "Qwen Image 2.0"
family = "qwen"
release_date = "2026-03-03"
last_updated = "2026-03-03"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false
[cost]
input = 0
output = 0
[limit]
context = 8_192
output = 0
[modalities]
input = ["text"]
output = ["image"]
@@ -0,0 +1,7 @@
base_model = "alibaba/qwen3.6-flash"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
@@ -0,0 +1,7 @@
base_model = "alibaba/qwen3.6-plus"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
@@ -0,0 +1,7 @@
base_model = "alibaba/qwen3.7-max"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
@@ -0,0 +1,7 @@
base_model = "alibaba/qwen3.7-plus"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
@@ -0,0 +1,20 @@
name = "Wan2.7 Image Pro"
release_date = "2026-05-29"
last_updated = "2026-05-29"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false
[cost]
input = 0
output = 0
[limit]
context = 8_192
output = 0
[modalities]
input = ["text"]
output = ["image"]
@@ -1,17 +1,20 @@
name = "Gemma 4 26B A4B Uncensored"
family = "gemma"
release_date = "2026-05-24"
last_updated = "2026-06-10"
name = "Wan2.7 Image"
release_date = "2026-05-29"
last_updated = "2026-05-29"
attachment = false
reasoning = false
temperature = true
tool_call = false
open_weights = false
[cost]
input = 0
output = 0
[limit]
context = 64_000
output = 4_096
context = 8_192
output = 0
[modalities]
input = ["text"]
output = ["text"]
output = ["image"]
@@ -0,0 +1,5 @@
name = "Alibaba Token Plan (China)"
env = ["ALIBABA_TOKEN_PLAN_API_KEY"]
npm = "@ai-sdk/openai-compatible"
doc = "https://www.alibabacloud.com/help/zh/model-studio/token-plan-overview"
api = "https://token-plan.cn-beijing.maas.aliyuncs.com/compatible-mode/v1"
@@ -0,0 +1,7 @@
base_model = "alibaba/qwen3.7-plus"
[cost]
input = 0
output = 0
cache_read = 0
cache_write = 0
+1
View File
@@ -4,6 +4,7 @@ release_date = "2025-10"
last_updated = "2025-10"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
knowledge = "2024-06"
tool_call = false
+6 -4
View File
@@ -1,20 +1,22 @@
name = "GPT OSS 120B"
family = "gpt-oss"
release_date = "2025-08-05"
last_updated = "2025-08-05"
last_updated = "2026-06-10"
attachment = false
reasoning = true
reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
temperature = true
tool_call = true
open_weights = true
structured_output = true
[cost]
input = 0.25
output = 0.69
input = 0.35
output = 0.75
[limit]
context = 131_072
output = 32_768
output = 40_960
[modalities]
input = ["text"]
@@ -1,23 +0,0 @@
name = "Llama 3.1 8B"
family = "llama"
release_date = "2025-01-01"
last_updated = "2026-05-27"
attachment = false
reasoning = false
temperature = true
knowledge = "2023-12"
tool_call = true
open_weights = true
status = "deprecated"
[cost]
input = 0.10
output = 0.10
[limit]
context = 32_000
output = 8000
[modalities]
input = ["text"]
output = ["text"]
+7 -4
View File
@@ -1,11 +1,14 @@
name = "Z.AI GLM-4.7"
release_date = "2026-01-10"
last_updated = "2026-01-10"
release_date = "2026-01-07"
last_updated = "2026-06-10"
attachment = false
reasoning = false
reasoning = true
reasoning_options = [{ type = "effort", values = ["none"] }]
temperature = true
tool_call = true
open_weights = true
structured_output = true
status = "beta"
[cost]
input = 2.25
@@ -15,7 +18,7 @@ cache_write = 0
[limit]
context = 131_072
output = 40_000
output = 40_960
[modalities]
input = ["text"]
@@ -4,6 +4,7 @@ release_date = "2025-12"
last_updated = "2026-02-25"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
knowledge = "2024-10"
@@ -4,6 +4,7 @@ release_date = "2026-02-12"
last_updated = "2026-02-25"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2025-12-01"
last_updated = "2025-12-12"
attachment = true
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
knowledge = "2025-12"
@@ -4,6 +4,7 @@ release_date = "2025-12"
last_updated = "2026-02-25"
attachment = true
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -1,4 +1,5 @@
base_model = "moonshotai/kimi-k2.6"
reasoning_options = []
[interleaved]
field = "reasoning_content"
@@ -4,6 +4,7 @@ release_date = "2025-08-05"
last_updated = "2026-02-25"
attachment = false
reasoning = true
reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2025-08-05"
last_updated = "2025-12-12"
attachment = false
reasoning = true
reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2025-07-31"
last_updated = "2026-02-25"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
structured_output = true
@@ -3,6 +3,7 @@ release_date = "2026-05-12"
last_updated = "2026-06-02"
attachment = true
reasoning = true
reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
tool_call = true
open_weights = false
knowledge = "2026-05"
@@ -0,0 +1,8 @@
base_model = "anthropic/claude-opus-4-8"
[cost]
input = 5.64
output = 28.198
cache_read = 0.563
cache_write = 7.049
@@ -9,6 +9,8 @@ knowledge = "2025-09"
tool_call = true
open_weights = true
reasoning_options = []
[interleaved]
field = "reasoning_content"
@@ -1,17 +1,16 @@
name = "Grok 4.20"
family = "grok"
release_date = "2026-03-12"
last_updated = "2026-06-10"
name = "Anthropic Claude Fable 5"
family = "claude-fable"
release_date = "2026-06-09"
last_updated = "2026-06-12"
attachment = true
reasoning = true
tool_call = true
structured_output = true
temperature = true
tool_call = true
open_weights = false
[limit]
context = 2000000
output = 128000
context = 1_000_000
output = 128_000
[modalities]
input = ["text", "image"]
@@ -0,0 +1,17 @@
name = "Nemotron 3 Ultra"
family = "nemotron"
release_date = "2026-06-04"
last_updated = "2026-06-12"
attachment = false
reasoning = false
temperature = true
tool_call = true
open_weights = false
[limit]
context = 131_072
output = 8_192
[modalities]
input = ["text"]
output = ["text"]
@@ -8,6 +8,9 @@ temperature = true
tool_call = true
open_weights = true
[[reasoning_options]]
type = "toggle"
[cost]
input = 0
output = 0
@@ -15,6 +15,10 @@ type = "toggle"
type = "effort"
values = ["low", "medium", "high"]
[[reasoning_options]]
type = "budget_tokens"
min = 1
[cost]
cache_read = 0.10
input = 0.50
@@ -2,6 +2,7 @@ name = "MiniMax-M2.5"
family = "minimax"
attachment = false
reasoning = true
reasoning_options = []
tool_call = true
structured_output = true
temperature = true
@@ -2,6 +2,7 @@ name = "GLM-5.1"
family = "glm"
attachment = false
reasoning = true
reasoning_options = []
tool_call = true
structured_output = true
temperature = true
@@ -2,6 +2,7 @@ name = "GLM-5"
family = "glm"
attachment = false
reasoning = true
reasoning_options = []
tool_call = true
structured_output = true
temperature = true
@@ -0,0 +1,11 @@
base_model = "anthropic/claude-fable-5"
[cost]
input = 10
output = 50
cache_read = 1
cache_write = 12.5
[limit]
context = 1_000_000
output = 128_000
@@ -4,6 +4,7 @@ release_date = "2025-08-28"
last_updated = "2025-08-28"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
structured_output = true
@@ -4,6 +4,7 @@ release_date = "2025-12-17"
last_updated = "2026-04-04"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
structured_output = true
@@ -4,6 +4,7 @@ release_date = "2025-11-13"
last_updated = "2025-11-13"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
structured_output = true
@@ -4,6 +4,7 @@ release_date = "2025-08-05"
last_updated = "2025-08-05"
attachment = false
reasoning = true
reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2025-08-05"
last_updated = "2025-08-05"
attachment = false
reasoning = true
reasoning_options = [{ type = "effort", values = ["low", "medium", "high"] }]
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2025-08-13"
last_updated = "2025-08-13"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
structured_output = true
@@ -4,6 +4,7 @@ release_date = "2026-01-06"
last_updated = "2026-01-06"
attachment = false
reasoning = true
reasoning_options = [{ type = "toggle" }]
temperature = true
tool_call = true
structured_output = true
@@ -4,6 +4,7 @@ release_date = "2026-02-11"
last_updated = "2026-02-11"
attachment = false
reasoning = true
reasoning_options = [{ type = "toggle" }]
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2026-02-12"
last_updated = "2026-06-01"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
structured_output = false
@@ -4,6 +4,7 @@ release_date = "2026-01-01"
last_updated = "2026-06-01"
attachment = false
reasoning = true
reasoning_options = [{ type = "toggle" }]
structured_output = true
temperature = false
tool_call = true
@@ -4,6 +4,7 @@ release_date = "2026-04-08"
last_updated = "2026-06-01"
attachment = false
reasoning = true
reasoning_options = [{ type = "toggle" }]
temperature = true
tool_call = true
structured_output = true
@@ -4,6 +4,7 @@ release_date = "2025-01-20"
last_updated = "2025-01-20"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
knowledge = "2024-12"
tool_call = true
+1
View File
@@ -4,6 +4,7 @@ release_date = "2024-12-01"
last_updated = "2025-11-13"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
knowledge = "2024-10"
tool_call = true
@@ -4,6 +4,7 @@ release_date = "2025-07-01"
last_updated = "2025-07-01"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
knowledge = "2025-04"
tool_call = true
+1
View File
@@ -4,6 +4,7 @@ release_date = "2024-12-01"
last_updated = "2024-12-01"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
knowledge = "2024-10"
tool_call = true
@@ -4,6 +4,7 @@ release_date = "2025-07-01"
last_updated = "2025-07-01"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
knowledge = "2024-12"
@@ -4,6 +4,7 @@ release_date = "2025-01-20"
last_updated = "2025-05-28"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
knowledge = "2024-07"
@@ -4,6 +4,7 @@ release_date = "2024-11-01"
last_updated = "2024-11-01"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
knowledge = "2024-08"
@@ -4,6 +4,7 @@ release_date = "2025-11"
last_updated = "2025-12"
attachment = false
reasoning = true
reasoning_options = []
structured_output = true
temperature = true
tool_call = true
@@ -9,6 +9,9 @@ tool_call = true
knowledge = "2025-04"
open_weights = true
[[reasoning_options]]
type = "toggle"
[interleaved]
field = "reasoning_content"
+1
View File
@@ -1,4 +1,5 @@
base_model = "alibaba/qwen3.6-35b-a3b"
reasoning_options = [{ type = "toggle" }]
[cost]
input = 5.0
@@ -4,6 +4,7 @@ release_date = "2025-12-23"
last_updated = "2025-12-23"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2026-02-13"
last_updated = "2026-02-13"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2026-02-12"
last_updated = "2026-02-12"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2026-03-18"
last_updated = "2026-03-18"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2026-03-18"
last_updated = "2026-03-18"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
open_weights = true
@@ -4,6 +4,7 @@ release_date = "2025-10-27"
last_updated = "2025-10-27"
attachment = false
reasoning = true
reasoning_options = []
temperature = true
tool_call = true
# knowledge = "2025-04" # Not listed on model page.

Some files were not shown because too many files have changed in this diff Show More