← variables.md

2026-10-10

174 change(s) in the model market

removed from the catalog

• Qwen: Qwen Plus 0728 (qwen/qwen-plus-2025-07-28) removed from the catalog
• Qwen: Qwen3 235B A22B (qwen/qwen3-235b-a22b) removed from the catalog
• Qwen: Qwen3 30B A3B Thinking 2507 (qwen/qwen3-30b-a3b-thinking-2507) removed from the catalog
• Qwen: Qwen3 8B (qwen/qwen3-8b) removed from the catalog
• Qwen: Qwen3 Coder Plus (qwen/qwen3-coder-plus) removed from the catalog
• Qwen: Qwen3 Max (qwen/qwen3-max) removed from the catalog
• Qwen: Qwen3 Max Thinking (qwen/qwen3-max-thinking) removed from the catalog
• Qwen: Qwen3 VL 235B A22B Thinking (qwen/qwen3-vl-235b-a22b-thinking) removed from the catalog
• Qwen: Qwen3 VL 32B Instruct (qwen/qwen3-vl-32b-instruct) removed from the catalog
• Qwen: Qwen3 VL 8B Thinking (qwen/qwen3-vl-8b-thinking) removed from the catalog
• Qwen: Qwen3.6 Max Preview (qwen/qwen3.6-max-preview) removed from the catalog

model price changes

• ~deepseek/deepseek-flash-latest: prompt price $0.016/M → $0/M (−99.4%)
• ~deepseek/deepseek-flash-latest: completion price $1.2/M → $0.6/M (−50.0%)
• ~deepseek/deepseek-pro-latest: prompt price $0.198/M → $0.3/M (+51.5%)
• ~deepseek/deepseek-pro-latest: completion price $8/M → $5/M (−37.5%)
• ~deepseek/deepseek-v4-flash-latest: prompt price $0.006/M → $0.01/M (+56.6%)
• ~deepseek/deepseek-v4-flash-latest: completion price $0.058/M → $1.28/M (+2117.8%)
• ~moonshotai/kimi-latest: prompt price $0.49/M → $0.33/M (−32.7%)
• ~moonshotai/kimi-latest: completion price $13/M → $14.9/M (+14.5%)
• ~z-ai/glm-flash-latest: prompt price $0.032/M → $0.04/M (+25.0%)
• ~z-ai/glm-flash-latest: completion price $0.553/M → $0.5/M (−9.7%)
• ~z-ai/glm-latest: prompt price $0.031/M → $0.039/M (+25.0%)
• ~z-ai/glm-latest: completion price $4.4/M → $4.8/M (+9.1%)
• deepseek/deepseek-v3.2: completion price $0.42/M → $0.8/M (+90.5%)
• deepseek/deepseek-v4-flash: prompt price $0.017/M → $0.018/M (+5.2%)
• deepseek/deepseek-v4-flash-0731: prompt price $0.008/M → $0.01/M (+25.3%)
• deepseek/deepseek-v4-pro: prompt price $0.955/M → $0.948/M (−0.7%)
• deepseek/deepseek-v4-pro: completion price $1.91/M → $1.9/M (−0.7%)
• google/gemma-4-26b-a4b-it: prompt price $0.09/M → $0.068/M (−25.0%)
• google/gemma-4-26b-a4b-it: completion price $0.3/M → $0.225/M (−25.0%)
• moonshotai/kimi-k2.5: prompt price $0.45/M → $0.5/M (+11.1%)
• moonshotai/kimi-k2.5: completion price $2.25/M → $2.5/M (+11.1%)
• moonshotai/kimi-k3: prompt price $0.5/M → $0.64/M (+28.0%)
• moonshotai/kimi-k3: completion price $12/M → $13.5/M (+12.5%)
• nvidia/nemotron-3.5-lightning: prompt price $0.047/M → $0.07/M (+49.3%)
• nvidia/nemotron-3.5-lightning: completion price $0.134/M → $0.2/M (+49.3%)
• qwen/qwen3-235b-a22b-thinking-2507: prompt price $0.23/M → $0.45/M (+95.7%)
• qwen/qwen3-235b-a22b-thinking-2507: completion price $2.3/M → $3.5/M (+52.2%)
• qwen/qwen3-vl-30b-a3b-thinking: prompt price $0.2/M → $0.29/M (+45.0%)
• qwen/qwen3-vl-30b-a3b-thinking: completion price $2.4/M → $1/M (−58.3%)
• qwen/qwen3-vl-8b-instruct: prompt price $0.117/M → $0.25/M (+113.7%)
• qwen/qwen3-vl-8b-instruct: completion price $0.455/M → $0.75/M (+64.8%)
• qwen/qwen3.6-27b: prompt price $0.45/M → $0.32/M (−28.9%)
• qwen/qwen3.6-27b: completion price $2.7/M → $3.2/M (+18.5%)
• z-ai/glm-4.6: prompt price $0.43/M → $0.5/M (+16.3%)
• z-ai/glm-4.6: completion price $1.75/M → $2/M (+14.3%)
• z-ai/glm-5.2: completion price $6/M → $7/M (+16.7%)

model context changes

• qwen/qwen3-14b: context length 131,072 → 40,960
• qwen/qwen3-235b-a22b-thinking-2507: context length 131,072 → 128,000
• qwen/qwen3-30b-a3b: context length 131,072 → 40,960

provider price changes

• SiliconFlow: deepseek/deepseek-v3.2 completion price $0.42/M → $0.8/M (+90.5%)
• Alibaba: deepseek/deepseek-v4-flash prompt price $0.134/M → $0.15/M (+11.9%)
• Alibaba: deepseek/deepseek-v4-flash completion price $0.268/M → $0.3/M (+11.9%)
• OpenInference: deepseek/deepseek-v4-flash prompt price $0.014/M → $0.019/M (+39.1%)
• OpenInference: deepseek/deepseek-v4-flash completion price $0.052/M → $1.15/M (+2117.8%)
• Relace: deepseek/deepseek-v4-flash prompt price $0.013/M → $0.018/M (+38.9%)
• Wafer: deepseek/deepseek-v4-flash prompt price $0.045/M → $0.075/M (+66.7%)
• Inceptron: deepseek/deepseek-v4-flash-0731 prompt price $0.05/M → $0.097/M (+94.8%)
• Inceptron: deepseek/deepseek-v4-flash-0731 completion price $0.65/M → $0.6/M (−7.7%)
• OpenInference: deepseek/deepseek-v4-flash-0731 prompt price $0.006/M → $0.01/M (+65.3%)
• OpenInference: deepseek/deepseek-v4-flash-0731 completion price $0.052/M → $0.27/M (+419.8%)
• Relace: deepseek/deepseek-v4-flash-0731 prompt price $0.006/M → $0.01/M (+65.0%)
• StreamLake: deepseek/deepseek-v4-flash-0731 prompt price $0.132/M → $0.088/M (−33.3%)
• StreamLake: deepseek/deepseek-v4-flash-0731 completion price $0.396/M → $0.264/M (−33.3%)
• Alibaba: deepseek/deepseek-v4-pro prompt price $1.42/M → $1.56/M (+10.2%)
• Alibaba: deepseek/deepseek-v4-pro completion price $2.83/M → $3.12/M (+10.2%)
• Reka: deepseek/deepseek-v4-pro prompt price $1.05/M → $1.2/M (+14.3%)
• Reka: deepseek/deepseek-v4-pro completion price $10.5/M → $12/M (+14.3%)
• StreamLake: deepseek/deepseek-v4-pro prompt price $0.955/M → $0.948/M (−0.7%)
• StreamLake: deepseek/deepseek-v4-pro completion price $1.91/M → $1.9/M (−0.7%)
• Sail Research: deepseek/deepseek-v4-pro-0813 completion price $4.3/M → $3/M (−30.2%)
• Sail Research: deepseek/deepseek-v4-pro-0813 completion price $4.3/M → $3/M (−30.2%)
• Wafer: deepseek/deepseek-v4-pro-0813 prompt price $0.22/M → $0.3/M (+36.4%)
• Wafer: deepseek/deepseek-v4-pro-0813 completion price $4/M → $5/M (+25.0%)
• DigitalOcean: deepseek/deepseek-v4.1-flash prompt price $0.225/M → $0.165/M (−26.7%)
• DigitalOcean: deepseek/deepseek-v4.1-flash completion price $0.9/M → $0.66/M (−26.7%)
• InferenceNet: deepseek/deepseek-v4.1-flash prompt price $0.05/M → $0.07/M (+40.0%)
• Ionstream: deepseek/deepseek-v4.1-flash prompt price $0.05/M → $0.1/M (+100.0%)
• Ionstream: deepseek/deepseek-v4.1-flash completion price $0.4/M → $1.1/M (+175.0%)
• Morph: deepseek/deepseek-v4.1-flash prompt price $0.058/M → $0.029/M (−50.0%)
• Morph: deepseek/deepseek-v4.1-flash completion price $0.6/M → $1/M (+66.7%)
• OpenInference: deepseek/deepseek-v4.1-flash prompt price $0.02/M → $0/M (−99.5%)
• OpenInference: deepseek/deepseek-v4.1-flash completion price $1/M → $0.36/M (−64.0%)
• Relace: deepseek/deepseek-v4.1-flash prompt price $0.016/M → $0/M (−99.4%)
• Relace: deepseek/deepseek-v4.1-flash completion price $1.2/M → $0.6/M (−50.0%)
• Wafer: deepseek/deepseek-v4.1-flash prompt price $0.045/M → $0.05/M (+11.1%)
• Wafer: deepseek/deepseek-v4.1-flash completion price $1.2/M → $1.6/M (+33.3%)
• NextBit: google/gemma-4-26b-a4b-it prompt price $0.09/M → $0.068/M (−25.0%)
• NextBit: google/gemma-4-26b-a4b-it completion price $0.3/M → $0.225/M (−25.0%)
• Io Net: mistralai/mistral-nemo prompt price $0.021/M → $0.018/M (−11.1%)
• Io Net: mistralai/mistral-nemo completion price $0.027/M → $0.024/M (−11.1%)
• SiliconFlow: moonshotai/kimi-k2.5 prompt price $0.45/M → $0.5/M (+11.1%)
• SiliconFlow: moonshotai/kimi-k2.5 completion price $2.25/M → $2.5/M (+11.1%)
• Decart: moonshotai/kimi-k3 prompt price $1.92/M → $1.98/M (+3.1%)
• Decart: moonshotai/kimi-k3 completion price $9.6/M → $9.9/M (+3.1%)
• InferenceNet: moonshotai/kimi-k3 prompt price $0.5/M → $0.64/M (+28.0%)
• InferenceNet: moonshotai/kimi-k3 completion price $12/M → $13.5/M (+12.5%)
• Morph: moonshotai/kimi-k3 prompt price $1.46/M → $0.339/M (−76.8%)
• Morph: moonshotai/kimi-k3 completion price $14/M → $14.9/M (+6.4%)
• Relace: moonshotai/kimi-k3 prompt price $0.49/M → $2/M (+308.2%)
• Relace: moonshotai/kimi-k3 completion price $13/M → $14/M (+7.7%)
• Wafer: moonshotai/kimi-k3 prompt price $0.5/M → $0.33/M (−34.0%)
• Wafer: moonshotai/kimi-k3 completion price $15/M → $14.9/M (−0.7%)
• Wafer: nvidia/nemotron-3.5-lightning prompt price $0.035/M → $0.045/M (+28.6%)
• Reka: qwen/qwen3.6-27b completion price $1.95/M → $2.5/M (+28.2%)
• ModelRun: qwen/qwen3.8-27b prompt price $0.8/M → $0.7/M (−12.5%)
• ModelRun: qwen/qwen3.8-27b completion price $5.25/M → $4.7/M (−10.5%)
• Wafer: qwen/qwen3.8-27b prompt price $0.045/M → $0.03/M (−33.3%)
• Io Net: xiaomi/mimo-v2.6-flash prompt price $0.29/M → $0.26/M (−10.3%)
• Io Net: xiaomi/mimo-v2.6-flash completion price $0.45/M → $0.41/M (−8.9%)
• Alibaba: z-ai/glm-5.2 prompt price $0.966/M → $1.12/M (+15.9%)
• Alibaba: z-ai/glm-5.2 completion price $3.04/M → $3.52/M (+15.9%)
• Cloudflare: z-ai/glm-5.2 prompt price $1.18/M → $0.5/M (−57.6%)
• Cloudflare: z-ai/glm-5.2 completion price $4.4/M → $6/M (+36.4%)
• Decart: z-ai/glm-5.2 prompt price $0.279/M → $0.273/M (−2.3%)
• Decart: z-ai/glm-5.2 completion price $1.72/M → $1.68/M (−2.3%)
• Decart: z-ai/glm-5.2 prompt price $2.75/M → $2.5/M (−9.1%)
• Decart: z-ai/glm-5.2 completion price $10/M → $9/M (−10.0%)
• Inceptron: z-ai/glm-5.2 prompt price $1.39/M → $1.25/M (−10.1%)
• Inceptron: z-ai/glm-5.2 completion price $4.39/M → $5.46/M (+24.4%)
• Morph: z-ai/glm-5.2 prompt price $0.278/M → $0.18/M (−35.3%)
• Morph: z-ai/glm-5.2 completion price $3.74/M → $6/M (+60.4%)
• StreamLake: z-ai/glm-5.2 prompt price $0.643/M → $0.556/M (−13.6%)
• StreamLake: z-ai/glm-5.2 completion price $2.02/M → $1.75/M (−13.6%)
• Wafer: z-ai/glm-5.2 completion price $6/M → $7/M (+16.7%)
• Wafer: z-ai/glm-5.2 completion price $8/M → $10/M (+25.0%)
• Inceptron: z-ai/glm-5.3 completion price $3.39/M → $2.2/M (−35.1%)
• Io Net: z-ai/glm-5.3 prompt price $1.2/M → $0.75/M (−37.5%)
• Io Net: z-ai/glm-5.3 completion price $4.4/M → $3.4/M (−22.7%)
• Morph: z-ai/glm-5.3 prompt price $0.214/M → $0.693/M (+223.8%)
• Morph: z-ai/glm-5.3 completion price $3.74/M → $6/M (+60.4%)
• Inceptron: z-ai/glm-5.3-flash prompt price $0.225/M → $0.12/M (−46.7%)
• Inceptron: z-ai/glm-5.3-flash completion price $0.6/M → $0.55/M (−8.3%)
• InferenceNet: z-ai/glm-5.3-flash prompt price $0.06/M → $0.07/M (+16.7%)
• Morph: z-ai/glm-5.3-flash prompt price $0.157/M → $0.117/M (−25.5%)
• Morph: z-ai/glm-5.3-flash completion price $0.7/M → $0.75/M (+7.1%)
• OpenInference: z-ai/glm-5.3-flash prompt price $0.032/M → $0.044/M (+37.5%)
• OpenInference: z-ai/glm-5.3-flash completion price $0.553/M → $0.45/M (−18.7%)
• StreamLake: z-ai/glm-5.3-flash prompt price $0.087/M → $0.069/M (−20.7%)
• StreamLake: z-ai/glm-5.3-flash completion price $0.29/M → $0.23/M (−20.7%)

providers added

• Alibaba now serves deepseek/deepseek-v4-flash-0731 (unknown, 1,000,000 ctx, $0.374/M in / $1.12/M out)
• Alibaba now serves deepseek/deepseek-v4-pro-0813 (unknown, 1,000,000 ctx, $1.32/M in / $3.96/M out)
• BaseTen now serves deepseek/deepseek-v4.1-flash (fp32, 1,048,576 ctx, $0.6/M in / $2.4/M out)
• Venice now serves openai/gpt-oss-120b (int4, 128,000 ctx, $0.03/M in / $0.15/M out)
• Tenstorrent now serves qwen/qwen3-32b (fp8, 20,000 ctx, $0.14/M in / $0.4/M out)
• Alibaba now serves qwen/qwen3.7-flash (unknown, 1,000,000 ctx, $0.034/M in / $0.15/M out)
• Alibaba now serves qwen/qwen3.7-max (unknown, 1,000,000 ctx, $2/M in / $6/M out)
• Alibaba now serves qwen/qwen3.7-plus (unknown, 1,000,000 ctx, $0.32/M in / $1.28/M out)
• Near AI now serves qwen/qwen3.8-27b (fp8, 262,144 ctx, $0.04/M in / $1.35/M out)
• Alibaba now serves qwen/qwen3.8-flash (unknown, 1,000,000 ctx, $0.15/M in / $0.47/M out)
• Alibaba now serves qwen/qwen3.8-max-0902 (unknown, 1,000,000 ctx, $2/M in / $6/M out)

providers removed

• Alibaba no longer serves deepseek/deepseek-v3.2 (was fp8)
• Novita no longer serves deepseek/deepseek-v4-flash (was fp8)
• Relace no longer serves deepseek/deepseek-v4-pro-0813 (was fp4)
• DeepInfra no longer serves inclusionai/ling-3.0-flash-fin (was fp4)
• Alibaba no longer serves qwen/qwen3-14b (was unknown)
• Alibaba no longer serves qwen/qwen3-235b-a22b-2507 (was unknown)
• StreamLake no longer serves qwen/qwen3-235b-a22b-2507 (was unknown)
• Alibaba no longer serves qwen/qwen3-235b-a22b-thinking-2507 (was unknown)
• Alibaba no longer serves qwen/qwen3-30b-a3b (was unknown)
• Alibaba no longer serves qwen/qwen3-30b-a3b-instruct-2507 (was unknown)
• StreamLake no longer serves qwen/qwen3-30b-a3b-instruct-2507 (was unknown)
• Alibaba no longer serves qwen/qwen3-coder (was unknown)
• Alibaba no longer serves qwen/qwen3-coder-30b-a3b-instruct (was unknown)
• Alibaba no longer serves qwen/qwen3-coder-next (was unknown)
• StreamLake no longer serves qwen/qwen3-coder-next (was unknown)
• Alibaba no longer serves qwen/qwen3-next-80b-a3b-instruct (was unknown)
• Alibaba no longer serves qwen/qwen3-next-80b-a3b-thinking (was unknown)
• Alibaba no longer serves qwen/qwen3-vl-235b-a22b-instruct (was unknown)
• Alibaba no longer serves qwen/qwen3-vl-30b-a3b-instruct (was unknown)
• Alibaba no longer serves qwen/qwen3-vl-30b-a3b-thinking (was unknown)
• Alibaba no longer serves qwen/qwen3-vl-8b-instruct (was unknown)
• Relace no longer serves z-ai/glm-5.2 (was unknown)
• Relace no longer serves z-ai/glm-5.3 (was unknown)

openrouter catalog snapshots 2026-10-09 → 2026-10-10 · schema 1 · generated 2026-10-10T11:11:13.101Z