https://benchr.org/ 2026-09-08T00:00:00Z The benchr team https://benchr.org/articles/kimi-k2-7-code-review 2026-09-08T00:00:00Z 2026-09-08T00:00:00Z

Moonshot

https://benchr.org/articles/deepseek-v4-flash-vision-exp-review 2026-09-07T00:00:00Z 2026-09-07T00:00:00Z

DeepSeek charges nothing extra for image input on this checkpoint. What you trade is stability: it is experimental and has no published license.

https://benchr.org/articles/gemini-3-8-flash-review 2026-09-07T00:00:00Z 2026-09-07T00:00:00Z

Gemini 3.8 Flash is priced to the cent like 3.7 Flash. Google says it spends more tokens on purpose, which is where the cost difference lives.

https://benchr.org/articles/claude-fable-5-1-migration 2026-09-02T00:00:00Z 2026-09-02T00:00:00Z

Claude Fable 5.1 keeps the $10/$50 API rate and 1M context, cuts cache reads to $0.25 per 1M tokens, and changes forced-tool and thinking behavior.

https://benchr.org/articles/deepseek-peak-pricing 2026-08-28T00:00:00Z 2026-08-28T00:00:00Z

DeepSeek raised V4 API prices on August 16, 2026 and split every rate into peak and off-peak halves. V4-Flash output went from $0.28 to $1.32 during peak hours.

https://benchr.org/articles/gemini-3-7-flash-price-parity 2026-08-24T00:00:00Z 2026-08-24T00:00:00Z

Gemini 3.7 Flash went GA on August 13, 2026 at $0.75/$3.75 per 1M tokens, the same rate 3.6 Flash was cut to. Both return to $1.50/$7.50 on January 1, 2027.

https://benchr.org/articles/gemini-3-7-flash-review 2026-08-21T00:00:00Z 2026-08-21T00:00:00Z

Google's stable multimodal model starts at $0.75/$3.75 through 2026, with a 1M-token window and a dated price change.

https://benchr.org/articles/glm-5-3-review 2026-08-21T00:00:00Z 2026-08-21T00:00:00Z

Z.AI's API release adds post-training gains, always-on reasoning, and three compatible protocols at $1.40/$4.40.

https://benchr.org/articles/grok-4-6-review 2026-08-21T00:00:00Z 2026-08-21T00:00:00Z

Grok 4.6 targets long coding agents, starts at $2/$6, doubles at 200K prompt tokens, and has no numeric text-output cap.

https://benchr.org/articles/kimi-k3-review 2026-08-21T00:00:00Z 2026-08-21T00:00:00Z

Moonshot's sparse agent model pairs a 1M context with flat hosted pricing and a deployment decision bigger than the token rate.

https://benchr.org/articles/deepseek-api-aliases-retired 2026-08-14T00:00:00Z 2026-08-14T00:00:00Z

DeepSeek retired deepseek-chat and deepseek-reasoner on July 24, 2026. Replace them with explicit V4 model IDs and validate before shipping.

https://benchr.org/articles/glm-5-2-review 2026-08-08T00:00:00Z 2026-08-08T00:00:00Z

GLM-5.2 review: Zhipu

https://benchr.org/articles/grok-4-5-review 2026-08-08T00:00:00Z 2026-08-08T00:00:00Z

Grok 4.5 review: xAI

https://benchr.org/articles/openai-presence-enterprise-agents 2026-08-04T00:00:00Z 2026-08-04T00:00:00Z

OpenAI Presence bundles voice and chat agents with policies, evaluations, approved actions, and escalation. What buyers know—and what remains undisclosed.

https://benchr.org/articles/claude-opus-5-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

The same $5/$25 list price, a 1M-token window, and one configuration change that can break a careless upgrade.

https://benchr.org/articles/gemini-3-1-flash-tts-preview-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Google

https://benchr.org/articles/gemini-3-5-flash-lite-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Google

https://benchr.org/articles/gemma-4-26b-a4b-it-launch 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Only 3.8B parameters activate per token, but all 25.2B weights stay loaded. Here is the deployment decision behind Gemma 4 26B A4B IT.

https://benchr.org/articles/gemma-4-31b-it-launch 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Gemma 4 31B IT review: dense architecture, official memory estimates, 256K context, sibling results, deployment fit, and when to avoid it.

https://benchr.org/articles/gpt-live-1-launch 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

OpenAI

https://benchr.org/articles/gpt-live-1-mini-launch 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

OpenAI

https://benchr.org/articles/gpt-realtime-2-1-mini-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

OpenAI

https://benchr.org/articles/gpt-realtime-2-1-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

OpenAI

https://benchr.org/articles/leanstral-1-5-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Mistral

https://benchr.org/articles/muse-image-launch 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

The image model is available in Meta AI, but the checked announcement gives developers no public model ID, rate card, or token limits.

https://benchr.org/articles/muse-video-preview-launch 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Meta previewed a video model with native audio alongside Muse Image, while leaving public developer terms unannounced.

https://benchr.org/articles/qwen-audio-3-0-tts-flash-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Alibaba

https://benchr.org/articles/qwen-audio-3-0-tts-plus-review 2026-07-28T00:00:00Z 2026-07-29T00:00:00Z

Alibaba

https://benchr.org/articles/gemini-3-6-flash-launch 2026-07-22T00:00:00Z 2026-08-24T00:00:00Z

Google released Gemini 3.6 Flash on July 21, 2026 at $1.50/$7.50, then halved both rates to $0.75/$3.75 through December 31, 2026.

https://benchr.org/articles/claude-sonnet-5-launch 2026-07-01T00:00:00Z 2026-08-31T00:00:00Z

Claude Sonnet 5 launched July 1, 2026 at $2/$10 per 1M — a rate Anthropic made standard on August 31, cancelling the planned September rise to $3/$15.

https://benchr.org/articles/gemini-3-5-pro-review 2026-07-01T00:00:00Z 2026-07-29T00:00:00Z

benchr previously published a review and pricing page for

https://benchr.org/articles/gpt-5-6-launch 2026-06-28T00:00:00Z 2026-08-24T00:00:00Z

GPT-5.6 (Sol, Terra, Luna) is generally available via the OpenAI API. Terra and Luna got cheaper on July 30, 2026, verified against OpenAI

https://benchr.org/articles/fine-tune-open-model-no-gpu 2026-06-19T00:00:00Z 2026-07-23T00:00:00Z

A practical QLoRA guide for fine-tuning 7B–70B open models on rented GPUs, with clearly labeled June 2026 planning scenarios.

https://benchr.org/articles/cuda-out-of-memory 2026-06-19T00:00:00Z 2026-06-19T00:00:00Z

The five real causes of the CUDA out of memory error when you serve an LLM, the fixes in order from free to last resort, and when you just need a bigger GPU.

https://benchr.org/articles/run-big-models-without-a-gpu 2026-06-19T00:00:00Z 2026-06-19T00:00:00Z

Why DeepSeek and Llama 70B won

https://benchr.org/articles/self-host-vs-api-gpu-cost 2026-06-19T00:00:00Z 2026-06-19T00:00:00Z

The honest break-even math for self-hosting an open model on a rented GPU versus a per-token API, and why utilization, not the sticker price, decides it.

https://benchr.org/articles/fable-mythos-export-control 2026-06-16T00:00:00Z 2026-07-30T00:00:00Z

The complete June 2026 Fable 5 and Mythos 5 export-control timeline: suspended June 12, controls lifted June 30, access restored July 1, and what developers…

https://benchr.org/articles/fable-5-vs-gpt-5-5-vs-gemini-3-1-pro 2026-06-13T00:00:00Z 2026-07-23T00:00:00Z

Claude Fable 5 vs GPT-5.5 vs Gemini 3.1 Pro: the three newest frontier models compared on price, context, published benchmarks, and the right model per use…

https://benchr.org/articles/which-claude-model 2026-06-13T00:00:00Z 2026-07-23T00:00:00Z

A decision guide to Anthropic

https://benchr.org/articles/opus-4-8-vs-gemini-3-1-pro 2026-06-13T00:00:00Z 2026-06-13T00:00:00Z

Claude Opus 4.8 vs Gemini 3.1 Pro: Opus leads coding and agentic work, Gemini is ~2.5x cheaper on input and edges abstract reasoning.