variables.md
the model market, diffed daily
Every day at 05:20 UTC this site snapshots the OpenRouter catalog — every model, every provider endpoint, prices, context windows, quantization labels — and diffs it against yesterday. What changed is published here as one-line facts. No commentary, no ranking.
Raw snapshots are retained from 2026-07-12 onward; every entry names the snapshot pair that produced it. The audience includes agents choosing a provider: every line is written to be quoted, and every page has a markdown twin.
Subscribe or poll:
rss ·
json feed ·
latest.json ·
latest (curl variables.md/changelog/latest speaks markdown)
2026-07-20 30 change(s)
removed from the catalog
- Venice: Uncensored (free) (cognitivecomputations/dolphin-mistral-24b-venice-edition:free) removed from the catalog
- Meta: Llama 3.2 3B Instruct (free) (meta-llama/llama-3.2-3b-instruct:free) removed from the catalog
- Meta: Llama 3.3 70B Instruct (free) (meta-llama/llama-3.3-70b-instruct:free) removed from the catalog
- Nous: Hermes 3 405B Instruct (free) (nousresearch/hermes-3-llama-3.1-405b:free) removed from the catalog
- Qwen: Qwen3 Coder 480B A35B (free) (qwen/qwen3-coder:free) removed from the catalog
- Qwen: Qwen3 Next 80B A3B Instruct (free) (qwen/qwen3-next-80b-a3b-instruct:free) removed from the catalog
model price changes
- nvidia/nemotron-3-super-120b-a12b: prompt price $0.21/M → $0.085/M (−59.5%)
- nvidia/nemotron-3-super-120b-a12b: completion price $0.455/M → $0.4/M (−12.1%)
- qwen/qwen3-14b: prompt price $0.227/M → $0.12/M (−47.3%)
- qwen/qwen3-14b: completion price $0.91/M → $0.24/M (−73.6%)
- qwen/qwen3-235b-a22b-thinking-2507: prompt price $0.15/M → $0.3/M (+100.7%)
- qwen/qwen3-235b-a22b-thinking-2507: completion price $1.5/M → $3/M (+100.7%)
- z-ai/glm-5: completion price $3.15/M → $2.55/M (−19.0%)
- z-ai/glm-5.2: prompt price $0.279/M → $0.958/M (+243.7%)
- z-ai/glm-5.2: completion price $0.876/M → $3.01/M (+243.7%)
model context changes
- z-ai/glm-5: context length 202,752 → 204,800
provider price changes
- Io Net: deepseek/deepseek-v4-flash prompt price $0.25/M → $0.239/M (−4.4%)
- Io Net: deepseek/deepseek-v4-flash completion price $0.399/M → $0.379/M (−5.0%)
- Io Net: qwen/qwen3.6-27b prompt price $0.376/M → $0.38/M (+1.2%)
- Io Net: qwen/qwen3.6-27b completion price $3.16/M → $3.2/M (+1.2%)
- Baidu: z-ai/glm-5.2 prompt price $0.28/M → $1.26/M (+350.0%)
- Baidu: z-ai/glm-5.2 completion price $0.88/M → $3.96/M (+350.0%)
- Decart: z-ai/glm-5.2 completion price $4.2/M → $4/M (−4.8%)
- Io Net: z-ai/glm-5.2 prompt price $2.34/M → $2.58/M (+10.0%)
- Io Net: z-ai/glm-5.2 completion price $5.12/M → $5.64/M (+10.0%)
- Novita: z-ai/glm-5.2 prompt price $0.276/M → $0.959/M (+247.7%)
- Novita: z-ai/glm-5.2 completion price $0.867/M → $3.01/M (+247.7%)
- StreamLake: z-ai/glm-5.2 prompt price $0.278/M → $0.958/M (+244.8%)
- StreamLake: z-ai/glm-5.2 completion price $0.873/M → $3.01/M (+244.8%)
provider context changes
- Io Net: qwen/qwen3.6-27b context length 262,140 → 32,768
2026-07-19 35 change(s)
model price changes
- google/gemma-4-26b-a4b-it: prompt price $0.1/M → $0.07/M (−30.0%)
- google/gemma-4-26b-a4b-it: completion price $0.3/M → $0.34/M (+13.3%)
- minimax/minimax-m2: prompt price $0.255/M → $0.3/M (+17.6%)
- minimax/minimax-m2: completion price $1.02/M → $1.2/M (+17.6%)
- moonshotai/kimi-k2.6: prompt price $0.95/M → $0.684/M (−28.0%)
- moonshotai/kimi-k2.6: completion price $4/M → $3.42/M (−14.5%)
- moonshotai/kimi-k2.7-code: prompt price $1/M → $0.85/M (−15.0%)
- moonshotai/kimi-k2.7-code: completion price $4.4/M → $3.8/M (−13.6%)
- qwen/qwen3-14b: prompt price $0.12/M → $0.227/M (+89.6%)
- qwen/qwen3-14b: completion price $0.24/M → $0.91/M (+279.2%)
- z-ai/glm-5.2: prompt price $0.343/M → $0.279/M (−18.8%)
- z-ai/glm-5.2: completion price $1.08/M → $0.876/M (−18.8%)
quantization changes
- Mancer 2: openai/gpt-oss-120b quantization fp4 → fp8
provider price changes
- Ambient: deepseek/deepseek-v4-flash prompt price $0.25/M → $0.5/M (+100.0%)
- Ambient: deepseek/deepseek-v4-flash completion price $0.4/M → $0.8/M (+100.0%)
- Ambient: moonshotai/kimi-k2.7-code prompt price $1/M → $0.85/M (−15.0%)
- Ambient: moonshotai/kimi-k2.7-code completion price $4.4/M → $3.8/M (−13.6%)
- Mancer 2: openai/gpt-oss-120b prompt price $0.04/M → $0.055/M (+37.5%)
- Mancer 2: openai/gpt-oss-120b completion price $0.35/M → $0.55/M (+57.1%)
- Io Net: qwen/qwen3.6-27b prompt price $0.341/M → $0.376/M (+10.0%)
- Io Net: qwen/qwen3.6-27b completion price $2.87/M → $3.16/M (+10.0%)
- Baidu: z-ai/glm-5.2 prompt price $0.42/M → $0.28/M (−33.3%)
- Baidu: z-ai/glm-5.2 completion price $1.32/M → $0.88/M (−33.3%)
- GMICloud: z-ai/glm-5.2 prompt price $0.98/M → $1.4/M (+42.9%)
- GMICloud: z-ai/glm-5.2 completion price $3.08/M → $4.4/M (+42.9%)
- Io Net: z-ai/glm-5.2 prompt price $2.13/M → $2.34/M (+10.0%)
- Io Net: z-ai/glm-5.2 completion price $4.66/M → $5.12/M (+10.0%)
- Novita: z-ai/glm-5.2 prompt price $0.343/M → $0.276/M (−19.6%)
- Novita: z-ai/glm-5.2 completion price $1.08/M → $0.867/M (−19.6%)
- StreamLake: z-ai/glm-5.2 prompt price $0.342/M → $0.278/M (−18.8%)
2026-07-18 48 change(s)
added to the catalog
- Auto Router (Beta) (openrouter/auto-beta) added to the catalog: 2,000,000 ctx, $-1000000/M in / $-1000000/M out
- Thinking Machines: Inkling (thinkingmachines/inkling) added to the catalog: 1,048,576 ctx, $1/M in / $4.05/M out
removed from the catalog
- Meta: Llama 3.2 11B Vision Instruct (meta-llama/llama-3.2-11b-vision-instruct) removed from the catalog
- NVIDIA: Llama 3.3 Nemotron Super 49B V1.5 (nvidia/llama-3.3-nemotron-super-49b-v1.5) removed from the catalog
model price changes
- mancer/weaver: prompt price $0.75/M → $0.5/M (−33.3%)
- mancer/weaver: completion price $1/M → $0.75/M (−25.0%)
- moonshotai/kimi-k2.7-code: prompt price $0.75/M → $1/M (+33.3%)
- moonshotai/kimi-k2.7-code: completion price $3.5/M → $4.4/M (+25.7%)
- qwen/qwen3-14b: prompt price $0.1/M → $0.12/M (+20.0%)
- qwen/qwen3-30b-a3b: prompt price $0.12/M → $0.13/M (+8.3%)
- qwen/qwen3-30b-a3b: completion price $0.5/M → $0.52/M (+4.0%)
- qwen/qwen3.5-27b: prompt price $0.195/M → $0.26/M (+33.3%)
- qwen/qwen3.5-27b: completion price $1.56/M → $2.6/M (+66.7%)
- z-ai/glm-4.7-flash: prompt price $0.06/M → $0.061/M (+0.8%)
- z-ai/glm-5.2: prompt price $0.967/M → $0.343/M (−64.5%)
- z-ai/glm-5.2: completion price $3.04/M → $1.08/M (−64.5%)
model context changes
- z-ai/glm-4.7-flash: context length 202,752 → 200,000
quantization changes
- Ambient: stepfun/step-3.7-flash quantization fp8 → fp4
provider price changes
- Mancer 2: gryphe/mythomax-l2-13b prompt price $0.5/M → $0.45/M (−10.0%)
- Mancer 2: gryphe/mythomax-l2-13b completion price $0.75/M → $0.65/M (−13.3%)
- Mancer 2: mancer/weaver prompt price $0.75/M → $0.5/M (−33.3%)
- Mancer 2: mancer/weaver completion price $1/M → $0.75/M (−25.0%)
- DekaLLM: mistralai/mistral-nemo prompt price $0.02/M → $0.018/M (−10.0%)
- Ambient: moonshotai/kimi-k2.7-code prompt price $0.75/M → $1/M (+33.3%)
- Ambient: moonshotai/kimi-k2.7-code completion price $3.5/M → $4.4/M (+25.7%)
- Mancer 2: openai/gpt-oss-120b prompt price $0.045/M → $0.04/M (−11.1%)
- Mancer 2: openai/gpt-oss-120b completion price $0.375/M → $0.35/M (−6.7%)
- Io Net: qwen/qwen3.6-27b prompt price $0.31/M → $0.341/M (+10.0%)
- Io Net: qwen/qwen3.6-27b completion price $2.61/M → $2.87/M (+10.0%)
- Mancer 2: undi95/remm-slerp-l2-13b prompt price $0.5/M → $0.45/M (−10.0%)
2026-07-17 36 change(s)
added to the catalog
- Meta: Muse Spark 1.1 (meta/muse-spark-1.1) added to the catalog: 1,048,576 ctx, $1.25/M in / $4.25/M out
- MoonshotAI: Kimi K3 (moonshotai/kimi-k3) added to the catalog: 1,048,576 ctx, $3/M in / $15/M out
model price changes
- ~moonshotai/kimi-latest: prompt price $0.66/M → $3/M (+354.5%)
- ~moonshotai/kimi-latest: completion price $3.41/M → $15/M (+339.9%)
- google/gemma-3-27b-it: prompt price $0.08/M → $0.1/M (+25.0%)
- google/gemma-3-27b-it: completion price $0.45/M → $0.3/M (−33.3%)
- meta-llama/llama-3.2-3b-instruct: prompt price $0.05/M → $0.051/M (+1.8%)
- meta-llama/llama-3.2-3b-instruct: completion price $0.33/M → $0.335/M (+1.5%)
- minimax/minimax-m2.7: prompt price $0.3/M → $0.25/M (−16.7%)
- minimax/minimax-m2.7: completion price $1.2/M → $1/M (−16.7%)
- mistralai/mistral-nemo: prompt price $0.02/M → $0.019/M (−5.0%)
- mistralai/mistral-nemo: completion price $0.04/M → $0.03/M (−25.0%)
- moonshotai/kimi-k2.7-code: prompt price $0.719/M → $0.75/M (+4.3%)
- moonshotai/kimi-k2.7-code: completion price $3.49/M → $3.5/M (+0.3%)
- qwen/qwen3-14b: prompt price $0.12/M → $0.1/M (−16.7%)
- qwen/qwen3.5-397b-a17b: prompt price $0.45/M → $0.39/M (−13.3%)
- qwen/qwen3.5-397b-a17b: completion price $3/M → $2.34/M (−22.0%)
- z-ai/glm-5.2: prompt price $0.969/M → $0.967/M (−0.1%)
- z-ai/glm-5.2: completion price $3.04/M → $3.04/M (−0.1%)
model context changes
- ~moonshotai/kimi-latest: context length 262,144 → 1,048,576
provider price changes
- DeepInfra: mistralai/mistral-nemo prompt price $0.02/M → $0.019/M (−5.0%)
- DeepInfra: mistralai/mistral-nemo completion price $0.04/M → $0.03/M (−25.0%)
- Ambient: moonshotai/kimi-k2.7-code prompt price $0.719/M → $0.75/M (+4.3%)
- Ambient: moonshotai/kimi-k2.7-code completion price $3.49/M → $3.5/M (+0.3%)
- Io Net: qwen/qwen3.6-27b prompt price $0.282/M → $0.31/M (+10.0%)
- Io Net: qwen/qwen3.6-27b completion price $2.38/M → $2.61/M (+10.0%)
- Io Net: z-ai/glm-5.2 prompt price $1.76/M → $1.94/M (+10.0%)
- Io Net: z-ai/glm-5.2 completion price $3.85/M → $4.24/M (+10.0%)
- Novita: z-ai/glm-5.2 prompt price $0.967/M → $0.965/M (−0.3%)
- Novita: z-ai/glm-5.2 completion price $3.04/M → $3.03/M (−0.3%)
2026-07-16 40 change(s)
removed from the catalog
- Arcee AI: Coder Large (arcee-ai/coder-large) removed from the catalog
model price changes
- google/gemma-4-31b-it: prompt price $0.12/M → $0.22/M (+83.3%)
- google/gemma-4-31b-it: completion price $0.37/M → $0.55/M (+48.6%)
- meta-llama/llama-3.2-3b-instruct: prompt price $0.051/M → $0.05/M (−1.8%)
- meta-llama/llama-3.2-3b-instruct: completion price $0.335/M → $0.33/M (−1.5%)
- moonshotai/kimi-k2.6: prompt price $0.66/M → $0.95/M (+43.9%)
- moonshotai/kimi-k2.6: completion price $3.41/M → $4/M (+17.3%)
- qwen/qwen3-14b: prompt price $0.1/M → $0.12/M (+20.0%)
- qwen/qwen3-coder-next: prompt price $0.12/M → $0.11/M (−8.3%)
- qwen/qwen3.7-max: prompt price $1.25/M → $1.48/M (+18.0%)
- qwen/qwen3.7-max: completion price $3.75/M → $4.42/M (+18.0%)
- z-ai/glm-4.7-flash: prompt price $0.061/M → $0.06/M (−0.8%)
- z-ai/glm-5.2: prompt price $0.895/M → $0.969/M (+8.3%)
- z-ai/glm-5.2: completion price $2.81/M → $3.04/M (+8.3%)
model context changes
- z-ai/glm-4.7-flash: context length 200,000 → 202,752
provider price changes
- Alibaba: deepseek/deepseek-v4-pro prompt price $1.61/M → $1.42/M (−11.9%)
- Alibaba: deepseek/deepseek-v4-pro completion price $3.22/M → $2.83/M (−11.9%)
- NextBit: google/gemma-4-26b-a4b-it prompt price $0.13/M → $0.12/M (−7.7%)
- NextBit: google/gemma-4-26b-a4b-it completion price $0.4/M → $0.35/M (−12.5%)
- DekaLLM: openai/gpt-oss-120b prompt price $0.036/M → $0.03/M (−16.7%)
- Io Net: qwen/qwen3.6-27b prompt price $0.345/M → $0.282/M (−18.2%)
- Io Net: qwen/qwen3.6-27b completion price $2.9/M → $2.38/M (−18.2%)
- Alibaba: qwen/qwen3.7-max prompt price $1.25/M → $1.48/M (+18.0%)
- Alibaba: qwen/qwen3.7-max completion price $3.75/M → $4.42/M (+18.0%)
- GMICloud: xiaomi/mimo-v2.5 prompt price $0.32/M → $0.112/M (−65.0%)
- GMICloud: xiaomi/mimo-v2.5 completion price $1.6/M → $0.224/M (−86.0%)
- GMICloud: xiaomi/mimo-v2.5-pro prompt price $0.8/M → $0.348/M (−56.5%)
- GMICloud: xiaomi/mimo-v2.5-pro completion price $2.4/M → $0.696/M (−71.0%)
- Io Net: z-ai/glm-5.2 prompt price $1.43/M → $1.76/M (+23.4%)
- Io Net: z-ai/glm-5.2 completion price $3.17/M → $3.85/M (+21.5%)
2026-07-15 86 change(s)
removed from the catalog
- Sao10K: Llama 3.1 70B Hanami x1 (sao10k/l3.1-70b-hanami-x1) removed from the catalog
model price changes
- deepseek/deepseek-chat-v3-0324: prompt price $0.24/M → $0.27/M (+12.5%)
- deepseek/deepseek-chat-v3-0324: completion price $0.9/M → $1.12/M (+24.4%)
- deepseek/deepseek-v3.1-terminus: completion price $0.95/M → $1/M (+5.3%)
- deepseek/deepseek-v3.2: prompt price $0.214/M → $0.269/M (+25.4%)
- deepseek/deepseek-v3.2: completion price $0.322/M → $0.4/M (+24.3%)
- deepseek/deepseek-v4-flash: prompt price $0.09/M → $0.098/M (+8.9%)
- deepseek/deepseek-v4-flash: completion price $0.18/M → $0.196/M (+8.9%)
- google/gemma-3-27b-it: completion price $0.16/M → $0.45/M (+181.2%)
- google/gemma-4-26b-a4b-it: prompt price $0.06/M → $0.1/M (+66.7%)
- google/gemma-4-26b-a4b-it: completion price $0.33/M → $0.3/M (−9.1%)
- google/gemma-4-31b-it: prompt price $0.06/M → $0.12/M (+100.0%)
- google/gemma-4-31b-it: completion price $0.35/M → $0.37/M (+5.7%)
- meta-llama/llama-3.1-8b-instruct: prompt price $0.02/M → $0.05/M (+150.0%)
- meta-llama/llama-3.1-8b-instruct: completion price $0.03/M → $0.08/M (+166.7%)
- meta-llama/llama-3.2-3b-instruct: prompt price $0.05/M → $0.051/M (+1.8%)
- meta-llama/llama-3.2-3b-instruct: completion price $0.33/M → $0.335/M (+1.5%)
- meta-llama/llama-3.3-70b-instruct: prompt price $0.1/M → $0.13/M (+30.0%)
- meta-llama/llama-3.3-70b-instruct: completion price $0.32/M → $0.4/M (+25.0%)
- minimax/minimax-m1: prompt price $0.4/M → $0.55/M (+37.5%)
- minimax/minimax-m2.7: prompt price $0.24/M → $0.3/M (+25.0%)
- minimax/minimax-m2.7: completion price $0.96/M → $1.2/M (+25.0%)
- mistralai/mistral-nemo: completion price $0.03/M → $0.04/M (+33.3%)
- mistralai/mistral-small-3.2-24b-instruct: prompt price $0.075/M → $0.1/M (+33.3%)
- mistralai/mistral-small-3.2-24b-instruct: completion price $0.2/M → $0.3/M (+50.0%)
- moonshotai/kimi-k2.5: prompt price $0.375/M → $0.57/M (+52.0%)
- moonshotai/kimi-k2.5: completion price $2.02/M → $2.85/M (+40.7%)
- nvidia/nemotron-3-super-120b-a12b: prompt price $0.08/M → $0.21/M (+162.5%)
- nvidia/nemotron-3-super-120b-a12b: completion price $0.45/M → $0.455/M (+1.1%)
- nvidia/nemotron-3-ultra-550b-a55b: prompt price $0.5/M → $0.6/M (+20.0%)
2026-07-14 100 change(s)
added to the catalog
- Kwaipilot: KAT-Coder-Air V2.5 (kwaipilot/kat-coder-air-v2.5) added to the catalog: 256,000 ctx, $0.15/M in / $0.6/M out
- Kwaipilot: KAT-Coder-Pro V2.5 (kwaipilot/kat-coder-pro-v2.5) added to the catalog: 256,000 ctx, $0.74/M in / $2.96/M out
removed from the catalog
- LiquidAI: LFM2.5-1.2B-Instruct (free) (liquid/lfm-2.5-1.2b-instruct:free) removed from the catalog
- LiquidAI: LFM2.5-1.2B-Thinking (free) (liquid/lfm-2.5-1.2b-thinking:free) removed from the catalog
- OpenAI: gpt-oss-120b (free) (openai/gpt-oss-120b:free) removed from the catalog
model price changes
- deepseek/deepseek-chat-v3.1: prompt price $0.21/M → $0.25/M (+19.0%)
- deepseek/deepseek-chat-v3.1: completion price $0.79/M → $0.95/M (+20.3%)
- deepseek/deepseek-v4-flash: prompt price $0.077/M → $0.09/M (+16.9%)
- deepseek/deepseek-v4-flash: completion price $0.154/M → $0.18/M (+16.9%)
- google/gemma-4-31b-it: prompt price $0.12/M → $0.06/M (−50.0%)
- moonshotai/kimi-k2.7-code: prompt price $0.72/M → $0.719/M (−0.1%)
- moonshotai/kimi-k2.7-code: completion price $3.5/M → $3.49/M (−0.3%)
- openai/gpt-oss-120b: prompt price $0.036/M → $0.03/M (−16.7%)
- openai/gpt-oss-120b: completion price $0.18/M → $0.15/M (−16.7%)
- qwen/qwen3.6-27b: prompt price $0.285/M → $0.289/M (+1.4%)
- z-ai/glm-5.2: prompt price $0.42/M → $0.928/M (+121.0%)
- z-ai/glm-5.2: completion price $1.32/M → $2.92/M (+121.0%)
provider price changes
- Amazon Bedrock: anthropic/claude-haiku-4.5 prompt price $1/M → $1.1/M (+10.0%)
- Amazon Bedrock: anthropic/claude-haiku-4.5 completion price $5/M → $5.5/M (+10.0%)
- Google: anthropic/claude-haiku-4.5 prompt price $1/M → $1.1/M (+10.0%)
- Google: anthropic/claude-haiku-4.5 completion price $5/M → $5.5/M (+10.0%)
- Amazon Bedrock: anthropic/claude-opus-4.5 prompt price $5/M → $5.5/M (+10.0%)
- Amazon Bedrock: anthropic/claude-opus-4.5 completion price $25/M → $27.5/M (+10.0%)
- Google: anthropic/claude-opus-4.6 prompt price $5/M → $5.5/M (+10.0%)
- Google: anthropic/claude-opus-4.6 completion price $25/M → $27.5/M (+10.0%)
- Amazon Bedrock: anthropic/claude-opus-4.7 prompt price $5/M → $5.5/M (+10.0%)
- Amazon Bedrock: anthropic/claude-opus-4.7 completion price $25/M → $27.5/M (+10.0%)
- Google: anthropic/claude-opus-4.7 prompt price $5/M → $5.5/M (+10.0%)
- Google: anthropic/claude-opus-4.7 completion price $25/M → $27.5/M (+10.0%)
- Amazon Bedrock: anthropic/claude-opus-4.8 prompt price $5/M → $5.5/M (+10.0%)
2026-07-13 2 change(s)
provider price changes
- Io Net: z-ai/glm-5.2 prompt price $1.96/M → $1.76/M (−10.0%)
- Io Net: z-ai/glm-5.2 completion price $4.35/M → $3.91/M (−10.0%)
$ curl variables.md — this site speaks markdown to your terminal.