← variables.md

2026-10-27

145 change(s) in the model market

added to the catalog

• OpenAI: GPT-6.1 Terra (openai/gpt-6.1-terra) added to the catalog: 1,050,000 ctx, $1/M in / $5/M out
• OpenAI: GPT-6.1 Terra Pro (openai/gpt-6.1-terra-pro) added to the catalog: 1,050,000 ctx, $1/M in / $5/M out
• OpenAI: GPT-6.1 Terra Pro (batch) (openai/gpt-6.1-terra-pro:batch) added to the catalog: 1,050,000 ctx, $0.5/M in / $2.5/M out
• OpenAI: GPT-6.1 Terra (batch) (openai/gpt-6.1-terra:batch) added to the catalog: 1,050,000 ctx, $0.5/M in / $2.5/M out

removed from the catalog

• Arcee AI: Trinity Large Thinking (arcee-ai/trinity-large-thinking) removed from the catalog
• IBM: Granite 4.0 Micro (ibm-granite/granite-4.0-h-micro) removed from the catalog
• Mistral: Codestral 2508 (batch) (mistralai/codestral-2508:batch) removed from the catalog
• Mistral: Ministral 3 8B 2512 (batch) (mistralai/ministral-8b-2512:batch) removed from the catalog
• Mistral: Mistral Large 3 2512 (batch) (mistralai/mistral-large-2512:batch) removed from the catalog
• Mistral: Mistral Medium 3.1 (batch) (mistralai/mistral-medium-3.1:batch) removed from the catalog
• Mistral: Mistral Small 4 (batch) (mistralai/mistral-small-2603:batch) removed from the catalog
• NVIDIA: Nemotron 3 Super (free) (nvidia/nemotron-3-super-120b-a12b:free) removed from the catalog

model price changes

• ~deepseek/deepseek-pro-latest: prompt price $0.257/M → $0.243/M (−5.3%)
• ~moonshotai/kimi-latest: prompt price $0.152/M → $0.179/M (+17.8%)
• ~moonshotai/kimi-latest: completion price $20/M → $11.5/M (−42.5%)
• ~openai/gpt-terra-latest: prompt price $2/M → $1/M (−50.0%)
• ~openai/gpt-terra-latest: completion price $12/M → $5/M (−58.3%)
• ~z-ai/glm-flash-latest: prompt price $0.041/M → $0.029/M (−31.0%)
• ~z-ai/glm-flash-latest: completion price $0.076/M → $0.092/M (+21.8%)
• deepseek/deepseek-v4-flash: prompt price $0.091/M → $0.05/M (−45.1%)
• deepseek/deepseek-v4-flash: completion price $0.182/M → $0.45/M (+147.3%)
• deepseek/deepseek-v4-flash-0731: prompt price $0.44/M → $0.119/M (−73.0%)
• deepseek/deepseek-v4-flash-0731: completion price $1.32/M → $0.238/M (−82.0%)
• deepseek/deepseek-v4.1-flash: prompt price $0.02/M → $0.14/M (+600.0%)
• deepseek/deepseek-v4.1-flash: completion price $1/M → $0.42/M (−58.0%)
• google/gemma-3-4b-it: prompt price $0.04/M → $0.05/M (+25.0%)
• google/gemma-3-4b-it: completion price $0.05/M → $0.1/M (+100.0%)
• moonshotai/kimi-k2.7-code: prompt price $0.671/M → $0.56/M (−16.6%)
• moonshotai/kimi-k2.7-code: completion price $3.35/M → $3.1/M (−7.5%)
• moonshotai/kimi-k3: prompt price $0.152/M → $0.179/M (+17.8%)
• moonshotai/kimi-k3: completion price $20/M → $11.5/M (−42.5%)
• qwen/qwen3.5-27b: prompt price $0.27/M → $0.26/M (−3.7%)
• qwen/qwen3.5-27b: completion price $2.16/M → $2.6/M (+20.4%)
• xiaomi/mimo-v2.5: prompt price $0.119/M → $0.4/M (+236.1%)
• xiaomi/mimo-v2.5: completion price $0.238/M → $2/M (+740.3%)
• xiaomi/mimo-v2.5-pro: prompt price $0.304/M → $0.48/M (+57.6%)
• xiaomi/mimo-v2.5-pro: completion price $0.609/M → $1.8/M (+195.6%)
• z-ai/glm-5.1: prompt price $1.4/M → $0.966/M (−31.0%)
• z-ai/glm-5.1: completion price $4.4/M → $3.04/M (−31.0%)

quantization changes

• Minimax: minimax/minimax-m2.1 quantization fp8 → unknown
• Minimax: minimax/minimax-m2.1 quantization fp8 → unknown
• Minimax: minimax/minimax-m2.5 quantization fp8 → unknown
• Minimax: minimax/minimax-m2.5 quantization fp8 → unknown
• Minimax: minimax/minimax-m2.7 quantization fp8 → unknown
• Minimax: minimax/minimax-m2.7 quantization fp8 → unknown
• Minimax: minimax/minimax-m3 quantization fp8 → unknown
• Moonshot AI: moonshotai/kimi-k2.6 quantization int4 → unknown
• Moonshot AI: moonshotai/kimi-k2.7-code quantization int4 → unknown
• Moonshot AI: moonshotai/kimi-k2.7-code quantization int4 → unknown
• Moonshot AI: moonshotai/kimi-k3 quantization mxfp4 → unknown
• Nebius: reflection/beam-1.0-501b-a23b quantization fp8 → fp4
• Modal: z-ai/glm-5.3-flash quantization nvfp4 → fp4

provider price changes

• Inceptron: deepseek/deepseek-v4-flash-0731 prompt price $0.11/M → $0.13/M (+18.2%)
• Wafer: deepseek/deepseek-v4-flash-0731 prompt price $0.14/M → $0.13/M (−7.1%)
• Relace: deepseek/deepseek-v4-pro-0813 prompt price $0.257/M → $0.243/M (−5.3%)
• CoreWeave: deepseek/deepseek-v4.1-flash prompt price $0.15/M → $0.13/M (−13.3%)
• DigitalOcean: deepseek/deepseek-v4.1-flash prompt price $0.165/M → $0.225/M (+36.4%)
• DigitalOcean: deepseek/deepseek-v4.1-flash completion price $0.66/M → $0.9/M (+36.4%)
• InferenceNet: deepseek/deepseek-v4.1-flash completion price $0.645/M → $0.6/M (−7.0%)
• InferenceNet: deepseek/deepseek-v4.1-flash completion price $0.645/M → $0.6/M (−7.0%)
• Morph: deepseek/deepseek-v4.1-flash prompt price $0.14/M → $0.034/M (−75.7%)
• Morph: deepseek/deepseek-v4.1-flash completion price $0.8/M → $0.6/M (−25.0%)
• OpenInference: deepseek/deepseek-v4.1-flash prompt price $0.02/M → $0.006/M (−68.1%)
• OpenInference: deepseek/deepseek-v4.1-flash completion price $1/M → $3.8/M (+279.7%)
• Decart: deepseek/deepseek-v4.1-flash-1017 prompt price $0.137/M → $0.129/M (−5.5%)
• Decart: deepseek/deepseek-v4.1-flash-1017 completion price $0.546/M → $0.516/M (−5.5%)
• InferenceNet: deepseek/deepseek-v4.1-flash-1017 completion price $0.5/M → $0.55/M (+10.0%)
• InferenceNet: deepseek/deepseek-v4.1-flash-1017 completion price $0.5/M → $0.55/M (+10.0%)
• Ionstream: deepseek/deepseek-v4.1-flash-1017 prompt price $0.14/M → $0.187/M (+33.6%)
• Ionstream: deepseek/deepseek-v4.1-flash-1017 completion price $0.39/M → $0.51/M (+30.8%)
• Morph: deepseek/deepseek-v4.1-flash-1017 prompt price $0.14/M → $0.034/M (−75.7%)
• Morph: deepseek/deepseek-v4.1-flash-1017 completion price $0.8/M → $0.6/M (−25.0%)
• Novita: deepseek/deepseek-v4.1-flash-1017 prompt price $0.24/M → $0.195/M (−18.7%)
• Novita: deepseek/deepseek-v4.1-flash-1017 completion price $0.96/M → $0.78/M (−18.7%)
• Phala: deepseek/deepseek-v4.1-flash-1017 prompt price $0.3/M → $0.2/M (−33.3%)
• StreamLake: deepseek/deepseek-v4.1-flash-1017 prompt price $0.21/M → $0.178/M (−15.0%)
• StreamLake: deepseek/deepseek-v4.1-flash-1017 completion price $0.84/M → $0.714/M (−15.0%)
• Wafer: deepseek/deepseek-v4.1-flash-1017 prompt price $0.13/M → $0.07/M (−46.2%)
• CoreWeave: ibm-granite/granite-4.2-8b prompt price $0.1/M → $0.06/M (−40.0%)
• InferenceNet: minimax/minimax-m3.1-flash prompt price $0.1/M → $0.15/M (+50.0%)
• InferenceNet: minimax/minimax-m3.1-flash completion price $0.55/M → $0.6/M (+9.1%)
• InferenceNet: minimax/minimax-m3.1-flash prompt price $0.1/M → $0.15/M (+50.0%)
• InferenceNet: minimax/minimax-m3.1-flash completion price $0.55/M → $0.6/M (+9.1%)
• SambaNova: minimax/minimax-m3.1-flash prompt price $0.3/M → $0.15/M (−50.0%)
• SambaNova: minimax/minimax-m3.1-flash completion price $1.2/M → $0.6/M (−50.0%)
• Chutes: moonshotai/kimi-k2.6 prompt price $0.45/M → $0.39/M (−13.3%)
• Chutes: moonshotai/kimi-k2.6 completion price $1.95/M → $1.8/M (−7.7%)
• Inceptron: moonshotai/kimi-k2.6 prompt price $0.63/M → $0.585/M (−7.1%)
• CoreWeave: moonshotai/kimi-k2.7-code prompt price $0.71/M → $0.6/M (−15.5%)
• CoreWeave: moonshotai/kimi-k2.7-code completion price $3.5/M → $3/M (−14.3%)
• Inceptron: moonshotai/kimi-k2.7-code prompt price $0.671/M → $0.56/M (−16.6%)
• Inceptron: moonshotai/kimi-k2.7-code completion price $3.35/M → $3.1/M (−7.5%)
• InferenceNet: moonshotai/kimi-k3 prompt price $2.24/M → $0.5/M (−77.7%)
• InferenceNet: moonshotai/kimi-k3 completion price $14.5/M → $11.9/M (−18.2%)
• Morph: moonshotai/kimi-k3 prompt price $1.17/M → $7.34/M (+527.4%)
• Morph: moonshotai/kimi-k3 completion price $18.8/M → $28.1/M (+49.8%)
• Relace: moonshotai/kimi-k3 prompt price $0.227/M → $0.18/M (−20.9%)
• Relace: moonshotai/kimi-k3 completion price $7.7/M → $9.72/M (+26.3%)
• Sail Research: moonshotai/kimi-k3 prompt price $1.2/M → $0.25/M (−79.2%)
• Sail Research: moonshotai/kimi-k3 completion price $5.5/M → $11/M (+100.0%)
• Sail Research: moonshotai/kimi-k3 prompt price $1.2/M → $0.23/M (−80.8%)
• Sail Research: moonshotai/kimi-k3 completion price $5.5/M → $11/M (+100.0%)
• Wafer: moonshotai/kimi-k3 prompt price $0.226/M → $0.179/M (−20.8%)
• Wafer: moonshotai/kimi-k3 completion price $20/M → $11.5/M (−42.5%)
• PrimeIntellect: qwen/qwen3.5-9b prompt price $0.1/M → $0.17/M (+70.0%)
• CoreWeave: qwen/qwen3.6-35b-a3b prompt price $0.25/M → $0.1/M (−60.0%)
• CoreWeave: qwen/qwen3.6-35b-a3b completion price $1.25/M → $0.95/M (−24.0%)
• DeepInfra: xiaomi/mimo-v2.6-flash prompt price $0.091/M → $0.14/M (+53.8%)
• DeepInfra: xiaomi/mimo-v2.6-flash completion price $0.182/M → $0.28/M (+53.8%)
• GMICloud: xiaomi/mimo-v2.6-pro prompt price $0.435/M → $0.37/M (−15.0%)
• GMICloud: xiaomi/mimo-v2.6-pro completion price $0.87/M → $0.739/M (−15.0%)
• Chutes: z-ai/glm-5.1 prompt price $0.98/M → $0.9/M (−8.2%)
• Chutes: z-ai/glm-5.1 completion price $3.08/M → $2.9/M (−5.8%)
• Morph: z-ai/glm-5.2 prompt price $0.73/M → $6.69/M (+815.9%)
• Morph: z-ai/glm-5.2 completion price $4.4/M → $8.14/M (+85.1%)
• Decart: z-ai/glm-5.3 completion price $2.71/M → $2.69/M (−0.7%)
• Inceptron: z-ai/glm-5.3 prompt price $0.6/M → $1.2/M (+100.0%)
• Inceptron: z-ai/glm-5.3 completion price $2.8/M → $3.5/M (+25.0%)
• Morph: z-ai/glm-5.3 prompt price $0.73/M → $6.69/M (+815.9%)
• Morph: z-ai/glm-5.3 completion price $4.4/M → $8.14/M (+85.1%)
• CoreWeave: z-ai/glm-5.3-flash prompt price $0.15/M → $0.12/M (−20.0%)
• CoreWeave: z-ai/glm-5.3-flash completion price $0.5/M → $0.38/M (−24.0%)
• Morph: z-ai/glm-5.3-flash prompt price $0.337/M → $0.281/M (−16.6%)
• OpenInference: z-ai/glm-5.3-flash prompt price $0.11/M → $0.03/M (−72.7%)
• OpenInference: z-ai/glm-5.3-flash completion price $0.01/M → $0.1/M (+900.0%)
• Reka: z-ai/glm-5.3-flash prompt price $0.15/M → $0.12/M (−20.0%)
• Reka: z-ai/glm-5.3-flash completion price $2.4/M → $0.5/M (−79.2%)
• Relace: z-ai/glm-5.3-flash prompt price $0.041/M → $0.029/M (−31.0%)
• Relace: z-ai/glm-5.3-flash completion price $0.076/M → $0.092/M (+21.8%)
• StreamLake: z-ai/glm-5.3-flash prompt price $0.043/M → $0.089/M (+103.4%)
• StreamLake: z-ai/glm-5.3-flash completion price $0.145/M → $0.295/M (+103.4%)

providers added

• Nebius now serves deepseek/deepseek-v4-pro-0813 (fp8, 979,000 ctx, $1.32/M in / $3.96/M out)
• PrimeIntellect now serves google/gemma-4-26b-a4b-it (bf16, 262,144 ctx, $0.1/M in / $0.3/M out)
• Wafer now serves moonshotai/kimi-k3 (mxfp4, 1,048,576 ctx, $0.5/M in / $22/M out)
• Nebius now serves nvidia/nemotron-3.5-super-vl (bf16, 1,048,576 ctx, $0.3/M in / $0.9/M out)
• Sail Research now serves openai/gpt-oss-120b (fp4, 131,072 ctx, $0.06/M in / $0.4/M out)
• Tenstorrent now serves qwen/qwen3.8-27b (fp4, 262,144 ctx, $0.06/M in / $0.5/M out)
• AtlasCloud now serves tencent/hy4-preview (fp8, 1,048,576 ctx, $0.834/M in / $2.5/M out)
• GMICloud now serves tencent/hy4-preview (bf16, 262,144 ctx, $0.834/M in / $2.5/M out)
• Io Net now serves tencent/hy4-preview (fp8, 262,124 ctx, $0.834/M in / $2.5/M out)
• Tenstorrent now serves z-ai/glm-5.3 (fp8, 1,044,480 ctx, $0.6/M in / $2.5/M out)

providers removed

• Reka no longer serves deepseek/deepseek-v4-pro (was unknown)
• DekaLLM no longer serves google/gemma-4-31b-it (was unknown)
• Wafer no longer serves minimax/minimax-m3.1-flash (was nvfp4)
• Io Net no longer serves reflection/beam-1.0-501b-a23b (was fp4)

openrouter catalog snapshots 2026-10-26 → 2026-10-27 · schema 1 · generated 2026-10-27T10:35:22.425Z